Barns H * Han pate
ye TOU re
ibe ret paren it a Neh
fi BAA be , te Pete
Adi eel
WA Pi shest Bete yeor erat)
hte Ritesh Me ;
SSS
i en; 9) ft i Pah t i itd Sibit i i { Aaa 3 AANRRR Eh Abt dis Heh igre haste Sieh ( i ue i Hae Ne Gul Ara tit f i fie i} ial CHD AE ; Peete AN f
ae i
tay test annie wiih ant at
ty
‘ tf
HH Garay ia
set
ae
Signs
oe peas
S323 = see SS
wee ste
” i Ane a
oo iH
Pete istst
rpadehe td i ; eee
EAs M07 BEATA ARE BM
AAA TAY
\
William Healey Dall .
! Division of Mollusks Sectional Library
See
TE EVEN
*
QL 420 .'t UG0OILTS Moall- MEMOIRS. CARNEGIE MUSEUM Vou. VIII. eta es . No. 3.
SOUTH AMERICAN NAIADES; A CONTRIBUTION TO THE KNOWLEDGE OF THE FRESHWATER MUSSELS OF SOUTH AMERICA.
By Dr. A. E. OrTMANN. bm *. Division of Molizcie
(Puares XXXIV-XLVIII) Zectional Library
INTRODUCTORY.
During the expedition of the Carnegie Museum to central South America, from 1907 to 1909, Mr. J. D. Haseman had as his prime object the collection of Fishes (Haseman, 1911)!. At the request of the present writer he, however, took particular pains to collect and preserve freshwater mussels. The result was one of the finest and largest collections of South American Nazades ever secured. The value of this collection is enhanced by the fact that a great number of specimens were preserved in alcohol with the soft parts, and the study of their anatomy has thus been made possible. Preliminary notes concerning the most important points of structure have been previously published (Ortmann, 1911a), chiefly in order to set forth the affinities of the South American forms with those of the rest of the world. Only a few typical forms were selected and discussed for this purpose. Further examination of the material has revealed a number of additional and highly interesting facts with regard to the anatomy, which throw light on the taxonomy and phylogeny of this group. 8
1 The references in parentheses are to the papers found in the bibliography at the end of this papgiges
451 ZReaenior
C 325.16 ( 2 \' Notional Wu
Instity, > %
JAN 09 1987 LIBRARIES
——
452 MEMOIRS OF THE CARNEGIE MUSEUM.
The present paper is intended to give a full account of these investigations. Naturally all accessible material has been included. In many cases only shells without their soft parts were available. This remark refers largely to older and some more recent material preserved in the Carnegie Museum obtained from various sources. The Haseman Collection likewise contains only the hard parts of many species, but on the other hand, other accessions consisted of shells with soft parts, as, for instance, the specimens of Anodontites crispatus from the-upper Magdalena- drainage in Colombia, collected by Dr. C. H. Eigenmann, and a collection of a number of species from the La Plata in Argentina, received from Dr. A. Wind- hausen.
In spite of the comparative wealth of material at hand, in some respects our collections are not complete and do not permit the deteemination of certain ques- tions, chiefly taxonomic. My greatest difficulty has been to properly identify the “species” at hand. The older writers, Spix, Wagner, Lea, D’Orbigny, Philippi, Hupé, and others, generally described their “species” from very insufficient ma- terial, and the author to whom we are most indebted for clearing up the taxonomy of the South American Naiades, H. Von Ihering, also was often handicapped by having too scanty material. All previous writers had no clear conception of the range of variation in the various forms, and thus their descriptions generally are those of individuals, often indeed very elaborate and complete, but without proper emphasis laid upon the really important specific characters. Simpson’s “ Descrip- tive Catalogue” (1914) did not much improve matters, since he was largely de- pendent upon the unsatisfactory publications of previous authors.
I do not claim, by any means, that my treatment of these shells overcomes all these difficulties; on the contrary, I am in many particulars not at all satisfied with the results obtained. Nevertheless, I claim that the study of the soft parts of a great many forms has furnished a basis for the proper understanding of what the “species” are, and has, at least, furnished a clue to their systematic arrange- ment, incomplete and fragmentary, it may be, but which probably will prove to be of great value, when the soft parts of all or most of the species are known.
The difficulty encountered in recognizing the described forms is sometimes exasperating. We should expect, in cases where exact type-localities are given, and where material from these is at hand, that it would not be very great. But, for instance, even of the species described by Von Ihering from Sao Paulo, I have recognized only a comparatively small number, although I possess a large quantity of material representing the Naiades from that state. In other cases, when the type-locality is vaguely, or not at all, given, it has been practically impossible to
ORTMANN: SOUTH AMERICAN NAIADES. 453
be sure of the identification. In consequence I have been compelled to introduce a number of “new species,” although I am afraid that some of them are not really “new.” But I must leave the task of making out their synonymy to others, who have access to authentic material representing the older forms. It should also be remembered that the introduction of new names is justified by the rules of nomenclature, when an original description is insufficient to enable the species to be recognized.
GENERAL REMARKS AS TO THE AFFINITIES AND GEOGRAPHICAL DISTRIBUTION OF THE SourH AMERICAN NATADES.
The earlier writers generally placed the South American forms in the old collective genera, Unio and Anodonta, to which a kind of intergrading group, called Monocondylea, and certain specialized types, such as Hyria and Castalia, were added. Von Ihering was the first to recognize that the South American ‘“ Ano- donta,” so-called, differs from the Anodonta of the northern hemisphere in important characters, and that it is related to certain African forms, Mutela, Spatha, ete. He calls this genus Glabaris Gray = Anodontites Bruguiére. But he left the other forms under Unio. For these Simpson (1900) used the name Diplodon Spix. In my preliminary report on South American Naiades (Ortmann, 191la, pp. 108, 120, 129, 130) I was able to show that Diplodon, as well as Hyria and Castalia, differ anatomically from the Unionide of the northern hemisphere, and that Von Ihering’s separation of Glabaris from Anodonta is fully justified and correct. I also found that the genera Hyria, Castalia (= Tetraplodon), and Diplodon, recognized by Simpson as a peculiar group, but still placed with the Unionidae, are actually more nearly allied to “Glabaris,” and form with this a group, the family Mutelide, which should be divided into two subfamilies, the Hyriine and Muteline, each witha number of genera, the anatomy of many of which, however, was still unknown.
Subsequent investigations brought out the fact that Simpson (1900) was cor- rect in associating certain Australian Naiades with the South American Diplodon, since I was able to show that D. australis has practically the same anatomical structure as the South American Diplodon (Ortmann, 1912), but that it should be elevated to the rank of a separate genus, Hyridella Swainson, admitted as a sub- genus by Simpson.
The systematic arrangement and geographical distribution of the families and subfamilies of the Naiades would thus be as follows:
454. MEMOIRS OF THE CARNEGIE MUSEUM.
Superfamily NATADES.
I. Family MARGARITANIDE......... Eurasia and North America (discontinuous).
II. Family Unronip: 1. Subfamily Unioninam.......Eurasia, Africa, North America, southward to Central America. 2. Subfamily ANODONTIN®”..... Eurasia, North America, southward to Central America. 3. Subfamily LAMPSILIN® ..... North America southward to Mexico.
III. Family Mure.ipa: 1. Subfamily Hyrimam........South America (but not in Central America), Australia. 2. Subfamily Muretin» ......South America to southern Mexico, Africa.
I can not improve upon this arrangement at present. It possibly might be advisable, in future, to elevate the two South American subfamilies to the rank of families, but I refrain from doing this, chiefly because the African and Australian forms belonging to them are too poorly known. The fact that in South America two groups of Naiades are found, which are more closely allied to each other than to any other group, is well expressed by uniting them into one family (Mutelide).
The more primitive subfamily, Hyriine, is found all over South America, but becomes rare in the northern parts (Venezuela and Colombia), and is missing, so far as known, in Central America. It is quite abundant in Chile, where the Mu- teline are absent, and it is also found in Australia.”
The more specialized group, subfamily Muteline, is found in South America, east of the Andes. It is missing in Chile, but has been reported from the Pacifie drainage in Ecuador. It goes into Central America and reaches southern Mexico. On the other hand it is represented in Africa, probably all over the continent, with the exception of the Mediterranean region.
This geographical distribution of the larger groups is extremely significant, for it serves to support certain theories as to the origin of the South American continent, its former connections, and the origin of its fauna. The presence of the Hyriine both in South America and Australia indicates the former connection of both continents, probably by way of Antarctica, and the fact that species of Diplodon, the most primitive of the South American Hyriine, are found in Chile, is entirely in keeping with this. The fact that Unionide with certam Hyrine structures® are found chiefly in southeastern Asia, suggests that they probably
2 It is unknown whether all the Australian Naiades belong here. Unionide related to the Hyriine in having the septa of the marsupium interrupted, are known to me from southeastern Asia (Siam, China) and northwestern America (Pacifie slope), but these forms certainly are Unionide in all other respects. .
‘ Interrupted septa in the marsupium are known in the Asiatie Unionine: Lamellidens (Ortmann, 191la, p. 106); in Hyriopsis and Contradens (Ortmann, Nautilus, XXX, 1916, pp. 85 and 106); and, according to the figures of Haas (System. Conchyl. Cabin., XIX, 1911 and 1912), also in Rectidens and
Acuticosta.
ORTMANN: SOUTH AMERICAN NAIADES. 455
originated on the old connecting land between Asia and Australia (Sino-Australian continent) or in Australia, and that Diplodon first reached Chile (in Mesozoic times) coming from Australia, and subsequently invaded the Brazilian mass. The modern forms of the Hyriinw (Castalia, Hyria) chiefly have their center in the basin of the Amazon, a comparatively young part of the continent.
The more advanced type, represented by the subfamily Muteline, developed probably in Brazil at the end of the Mesozoic, when it had a chance to spread over the old connection across the Atlantie (Archhelenis of Von Ihering), and the immigration of this stock into Central America is of rather late date (late Tertiary). It is not very likely that the Muteline reached South America coming from Africa, because there is no trace of them found in southern Asia.’
Dracnostic CHARACTERS OF THE SourH AMERICAN NAIADEs. (See Text-figures 1, 2, 3.) . Fo ’ Family: MUTELID Ortmann (1911).°
1. Diaphragm between branchial and cloacal cavity formed anteriorly by the gills, posteriorly by a solid union of the mantle margins (Figs. 1, 2, 3, t). (No such mantle connection in Unionidae).
2. Anterior end of inner gill (Figs. 1, 2, 3, i) broadly attached, and in contact with posterior base of palpi (h). (In the Union- ide, there is always a longer or shorter space between these parts.)
3. Anal and branchial openings (a and Fig. 1. Diagram of soft parts of female
b) sharply separated from each other by the of Diplodon trifidus (Lea). Natural size, union of the mantle margins (See Character left section of mantle removed. a Anal 1 1 i | 1 al but opening; 6, Branchial opening; h, Palpi; 7, Je anal opening Open or Closed above, Dut Inner gill; 0, Outer gill; p, Pes: s, Closed
there never is a supra-anal opening. (In the part of anal opening; ¢, Union of mantle Unionidae, the anal may be open, but, when _ separating anal and branchial openings. closed, there is always a supra-anal opening.)
4Tn Africa there are also Unionide of the subfamily Unionine; they undoubtedly point to a con- nection with Asia, since such forms are plentiful there, and this indicates a different route of immigra- tion from that of the Muteline.
° The name depends upon the investigation of the anatomy of the African genus Mutela, which is unknown, but we have every reason to assume, chiefly through Von Ihering’s study of the shell, that this genus will fall under this family. The family name Mutelid@ was first used by H. & A. Adams (1858), but for an entirely different association of forms. The Mutelide of Simpson (1900) correspond to our Muteline.
456 MEMOIRS OF THE CARNEGIE MUSEUM.
4. Gills with or without water-tubes, and with isolated, scattered inter- laminar connections, or with interrupted or solid septa. (In the Unionidae, mostly uninterrupted septa and water-tubes are found; very rarely are they interrupted; the forms with the latter probably form the connection with the Mutelide.)
5. Marsupium only in the inner gill (Figs. 1, 2, 3, i). (In the Unionidae, the marsupium is either in all four gills, or in the outer gills, never in the inner gills alone.)
6. Certain advanced genera of the Mutelide show a tendency to close the branchial opening in front by a mantle connection. (Such a connection is entirely unknown in the Unionidae.)
The foregoing are the anatomical characters. The shape of the shell and its parts are subject to so many variations, both in the Mutelide and the Unionide, that it is practically impossible to point out general differentiating characters. Different and often very peculiar types of shell, occurring independently, are fre- quently observed, so that it is clear that the shells by responding to certain stimuli and requirements have acquired sim- ilar shapes in forms, which belong to different families (convergency or parallelism). Simpson (1900 and 1914) found it impossible to give shell-characters for his subfamily “ Hyriane”’ = Hyriine+certain Uni- onine, except beak-sculpture, and he was mistaken in this. In our arrange- ment, this is the only character, which might be mentioned, but not
without qualification. We may ex-
MG. 2 iagram es arts omi f : . Fig. 2. Diagram of the soft parts of female of press it thus: The Mutelide have ra-
Caslalina nehringi Von Ihering. Natural Size, left . 5 3 dial beak-sculpture, if such is present
section of mantle removed. (Lettering as in Text- figure 1, p. 455.) at all);° while the Unionide rarely have distinct radial beak-sculpture, ‘but commonly other types, zig-zag, double-looped, or concentric. Some Mutelide have, indeed, shells which completely mimic those of certain Unionide. Species of Diplodon often externally So much resemble certain species ® But even this should be qualified. In certain species of Anodontites in the subfamily Muteline I have observed something like concentric beak-sculpture, while, as a rule, the Muteline have no beak- sculpture whatever.
ORTMANN: SOUTH AMERICAN NAIADES. 457
of the North American Elliptio, that without a minute examination of muscular impressions, hinge-teeth, etc., it is impossible to recognize them, when the locality is unknown. The only reliable character in this case, beak-sculpture, is often obliterated by erosion. Species of Anodontites often look like species of Anodonta. Here the examination of beak-sculpture, muscular impressions, and the ligamentinal sinus establishes their affinity. On the other hand, there are Mutelide which are easily recognized by the peculiar shape of the shell, and could never be confounded with Unionide (Castalia, H yria, Mycetopoda, ete.). The characters of the two subfamilies of the Mutelide are the following:
Subfamily Hyruna: Ortmann (1911).
Hyriane Swarnson (1840), very cloself corresponds to this. Lamphoramphus-
group of Hyrianew Simpson (1900).
1. Anal opening (Figs. 1, 2, a) closed above (s), slit-like (in South American forms), or forming a short, tubular siphon (Australian forms).
2. Marsupium generally an interrupted network of interlaminar connections (Figs. 1, 2, 1), the connections often standing in rows, thus forming incomplete septa and incomplete, communicating, water-tubes. In rare cases, the inter- laminar connections of the marsupium (but not of the non-marsupial gills) form solid septa and isolated water-tubes. Marsupial part of gill often restricted to only a section of it.
3. Non-marsupial gills always with poorly developed, scattered interlaminar connections. (Figs. 1, 2, 0.)
4, Inner lamina of inner gills always entirely connected with abdominal sac.
5. Palpi (Figs. 1, 2, h) subtriangular or subfalciform, with gently curved lower margins, and somewhat produced posterior points.
6. Larva a glochidium (See Fig. 4, p. 469).
As a character of the shell may be mentioned the beak-sculpture, which is generally present (often poorly developed, rarely absent), and always radial. The hinge-teeth are always present, and generally well-developed. Dorsal muscle- scars are present.
Subfamily Mure.ina Ortmann (1911).
Mutelide ApaAms (1858), Stmpson (1900). The name depends, of course, on the genus Mutela, the anatomy of the soft parts of which, as has been pointed out already, remains to be investigated.
1. Anal opening (Fig. 3, a) open (in South American forms, except Myceto- poda) or closed above (in African forms and Mycetopoda), slit-like.
458 MEMOIRS OF THE CARNEGIE MUSEUM.
2. Marsupium (Fig. 3, 7) with well-developed, continuous septa, forming well- defined, isolated water-tubes, with a peculiar longitudinal ridge or swelling on the septa near the outer lamina of the gill (Pl. XLVIII, figs. 6, 8). When gravid, only the inner compartment of the water-tubes (towards the inner lamina) is some- what extended and filled with eggs; the outer compartment becomes a secondary water-tube (Pl. XLVITI, fig. 7b).
3. Non-marsupial gills (Fig. 3, 0) also with septa and water-tubes, but the septa less strongly developed and without a ridge.
4. Inner lamina of inner gills entirely connected with abdominal sac (South Ameri- can forms), or free from it (African forms).
5. Palpi nearly semicircular, longer than high, with a short posterior truncation, strongly curved lower margins, and indis- tinct posterior points (Fig. 3, h).
6. Larva supposed to be a lasidium
tts _ (aecording to Von Ihering).
Fic. 3. Diagram of soft parts of female of Anodontites patagonica rubicunda (Lea). Nat- The shell generally has no beak-seulp- ural size, left section of mantle removed. a, ture whatever. In very rare cases a trace Anal opening; b, Branchial opening; h, Palpi; of eoneentric sculpture has been observed
i, Inner gill; 0, Outer gill; p, Pes; t, Union of (see ynder Anodontites trapezea). Hinge-
mantle separating anal and branchial openings. seethnaroncomless Gbeolese! Keduecd tennis ber, or size, and very often entirely absent (Anodontine type of hinge). Dorsal muscle-scars mostly absent in South American forms, very rarely a faint trace seen, or a few are present, as in Leila. In African forms, there is one single well-
developed dorsal muscle-scar.
Subfamily HYRIIN 42 Ortmann. GENERAL REMARKS.
Shell of various shapes, subelliptical, subtrapezoidal, subovate, suborbicular, or subtriangular, sometimes more or less alate. Beak-sculpture mostly present, rarely missing, but often indistinct or obliterated by erosion, always of the radial type, with two sets of radial ridges, starting from two points, immediately in front and immediately behind the tip of the umbo, and extending to a varying degree upon the disk. The posterior ridges of the anterior set, and the anterior ridges of the posterior set, generally interfere with each other in the middle of the disk,
=
ORTMANN: SOUTH AMERICAN NAIADES. 459
coming there into contact in a sharp angle. There are sometimes irregularities in the beak-sculpture obscuring their radial character.
Hinge-teeth always present, but rather variable in number, size, and shape. Normally, there are, as in most other Naiades, two laterals in the left valve, and one lateral in the right valve, but individual variations may occur in this respect. The pseudo-cardinals are much more variable. Originally their arrangement seems to be similar to that found in primitive Unionidae: that is to say, there are two in the left valve, and one or three in the right; of the latter, the middle one may be the main tooth, fitting in between the two teeth of the left valve, and the anterior and posterior tooth may be accessory. However, this arrangement is very often changed by the suppression of certain teeth, and the addition of others. The most general condition is when of the two pseudocardinals of the left valve the posterior is more or less obsolete, and often entirely wanting, and only the two anterior teeth of the right valve are present. Thus the left valve appears to have only one pseudocardinal (Aspidon of Von Ihering) and two laterals, and the right valve has two pseudoecardinals (Dexion and Epidexion), and one lateral. This is chiefly the case in certain species of Diplodon, where the pseudocardinals are greatly compressed, and directed more or less parallel to the hinge-line, and this should be kept in mind for the distinction of those species of Diplodon, which resemble in shape certain North American species of Elliptio, where the normal arrangement shows two pseudocardinals in the left, and one in the right valve.
This isnot the only variation. Additional teeth may turn up, and chiefly when in the left valve anterior to (or above) the anterior pseudocardinal an additional tooth is developed (Epaspidon, Von Thering).
The names Aspidon and Epaspidon, Dexion and Epidexion, may be used to advantage; but we should not forget that originally there is in the left valve a tooth behind the Aspidon (regarded by Von Ihering as being accessory), and that there is often in the right valve another (third) posterior tooth, behind the Dexion.’
Further, there is much variability in the shape of the teeth. The pseudo- cardinals, as has been stated, are often compressed, lamellar, and smooth. But often they are more solid, or stumpy, when they are frequently more or less split and divided, sometimes almost cut up into a number of teeth. The laterals may be smooth or corrugated, the corrugations standing obliquely on the faces; and in certain genera there are characteristic regular and parallel striations or ridges
7 Compare also Odhner, 1918, p. 574 ff. (Homologies of hinge-teeth of “ Unionidae ”). Odhner
seems to be right in a general way, but the hinge of the Navades is extremely complex and individually variable.
460 MEMOIRS OF THE CARNEGIE MUSEUM.
standing vertically to the edge of the teeth. Similar ridges may be present on the pseudocardinals.
In the shape of the muscle-sears on the inside of the shell, there exists no great difference between the Hyriinw and Muteline. The general rule is in the Hyriine that the upper anterior retractor-scar is distinctly separated from the ante- rior adductor impression, while the lower anterior retractor-scar is confluent with the latter. Very rarely the lower anterior retractor-scar is isolated. (In the Muteline either all three sears are confluent, or the lower anterior retractor-scar is isolated; the upper anterior retractor-scar is hardly ever isolated). |The posterior muscle-scars are confluent, the posterior retractor-sear forming a small process at the upper margin of the adductor-sear. The muscle-scars may be deeper or shallower, the upper an- terior retractor-sear generally is rounded, small, and remarkably deep.
Dorsal muscle-sears are present in the Hyrtine (mostly absent in the Muteline) ; they are variable in number, generally only a few, and are located in the bottom of the shallow beak-cavity, forming an irregular row parallel to the hinge-line, or somewhat oblique. In some forms with deep beak-cavities, these scars are situated on the inner side of the hinge-plate (behind the pseudocardinals and the inter- dentum).
The line of the mantle-impression is always simple, without a sinus poster- iorly. The ligamental sinus lies on the margin of the shell behind the ligament, over the posterior part of the lateral teeth (rarely more in front), and is always small and shallow (longer than deep). (in the Mutelinew, the ligamental sinus is generally much deeper, with a sharp lower point.)
The characters of the soft parts of the Hyriine have been pointed out above.
THE GENERA OF THE HyrIIN«&.
The genera of the Hyriine have been hitherto differentiated according to the shape of the shell, the character of the beak-sculpture, and the character and the sculpture of the hinge-teeth. According to Simpson (1914 pp. 1194 ff.) there are seven of them in South America: Tetraplodon Spix (recte Castalia Lamarck); Castalina Von Ihering; Castaliella Simpson; Callonaia Simpson; Hyria Lamarck; Prisodon Schumacher; Diplodon Spix.
In addition, there are species belonging to this group in Australia (and probably also in New Zealand), which have been placed by Simpson in Diplodon, but have been separated by him under the subgenus Hyridella Swainson. I have shown (Ortmann, 1912) that the type of Hyridella (Unio australis Lamarck) actually is closely allied to Diplodon, but that probably it is better to regard Hyridella as a genus by itself.
ORTMANN: SOUTH AMERICAN NAIADES. 461 The differences of the South American genera may be tabulated as follows:
a. Shell not alate, or only very slightlyalate behind, subelliptical, subovate, subrotund, or subtriangular. b. Shell subelliptical, subovate, or subrotund, but not subtriangular. Beaks not much elevated, with radial beak-seulpture, variously developed. Posterior ridge absent or poorly developed; when present and distinet, the shell is elongated. Hinge teeth more or less smooth or dis- sected, or corrugated, but without vertical parallel ridges. Interdentum very narrow. Diplodon. 6b. Shell subtriangular (or subquadrate). Beaks more or less elevated and with rather deep beak- cavities. With or without beak-sculpture. Posterior ridge well developed, defining a distinct posterior slope. Hinge-teeth generally much dissected, and often with parallel vertical ridges. Interdentum well developed, rather broad. c. Sides of shell somewhat flattened, posterior ridge moderate, posterior slope elevated in the middle and slightly alate (thus the shell becomes subalate). Hinge-teeth smooth, erenu- iigeiel rove yaa ompeyence lel Mate eee bain cit cece Sein ae ean ae ere CUI (108 ce. Sides of shell more or less convex, posterior ridge sharp, posterior slope truncated, not elevated (or only so in the young). Hinge-teeth with parallel ridges. d. Radial beak-seulpture present. e. Beak-seulpture strongly developed, extending over a large part of the disk. Surface MommCONCenthicalliy sul cAbeC sam ere patel u- ts ciierieie «cho eye versus 8 oe waists Castalia.
ee. Beak-seulpture short anteriorly, extending farther on the disk posteriorly. Surface
covered with strong, even, concentric ridges (suleated).............Castaliella. dd. Beaks without distinct sculpture, and surface of shell smooth.............Callonaia. aa. Shell subrhomboidal or subtriangular, strongly alate behind, slightly alate in front. b. Beaks with strong radial sculpture. Posterior ridge moderately developed.............Hyria. bb. Beaks without sculpture. Posterior ridge strongly developed, sharp and high... ... . . Prisodon.
From the expressions used in distinguishing these genera, it appears that Diplodon is a rather composite genus, containing shells of various, more or less indifferent types, while the other genera represent more marked and _ peculiar shapes, with Castalina forming, to a degree, a connection between Diplodon and the rest. This indicates that Diplodon includes the more primitive and generalized forms of the subfamily, while the others are more specialized and advanced. In a general way, this is supported by the investigation of the anatomy. However, it is much to be regretted that I am unable to give any information as to the struc- ture of the soft parts of the genera Castaliella and Callonaia. The former is en- tirely unknown to me, and of the latter I possess only two odd valves. Also of Prisodon I only have the soft parts of a young individual. Better material is at hand to represent the other genera.
Generally speaking, the anatomy of all these forms is rather uniform; yet there are certain characters in which they differ, and some of these, indeed, indi- cate a gradual advance in the structure.
462 MEMOIRS OF THE CARNEGIE MUSEUM.
One of these characters has been noticed long ago, and has been discussed in detail by Von Ihering. This is the tendency manifested in certain forms to close the branchial opening in front by a connection of the mantle-margins, which is a character entirely wanting in the Unionidae, but it is found occasionally in both subfamilies of the Mutelide (Hyriine and Muteline). But numerous members of this family do not have this character, and thus it is evident that the absence of this mantle-connection indicates a primitive condition, while its presence is a more advanced stage, expressing the tendency to transform the branchial opening into a closed tube (siphon).
According to Von Ihering (1898), the mantle-connection closing the branchial opening anteriorly is always present in Castalia. I have been able to confirm this only partially. Of typical species of Castalia I possess the soft parts of six specimens of C. acuticosta: they show the branchial opening closed with one ex- ception, where it is open. In C. wndosa, a somewhat aberrant type, Von Ihering describes the branchial as normally closed, but he says that in four out of twenty- one specimens examined the mantle-connection was missing.* Of this species I have five specimens with soft parts, of which three have the branchial closed, while it is open in two of them. In one of the latter, however, a small one, it may be torn. It must be admitted that also in other cases this connection of the mantle may have been torn in life or in preservation.’ And thus it might be that normally this mantle-connection is present in Castalia.
Castalina, according to Von Ihering, varies in the development of this character in the different species and individuals. He has investigated two males of C. nehringi, and one had the branchial closed, the other open. In eight females ex- amined by myself, I found the branchial open. In C. martensi Von Ihering found the branchial opening in most cases closed. I have no soft parts of this species. But in two specimens of C. psammoica, which I possess, one has this opening dis- tinetly closed in front, in the other this is not the case.
Nowhere else within this subfamily has this character been found. Castaliella and Callonaia may possess it, but their anatomy isunknown. Of Hyria and Prisodon
8 In Von Ihering’s description (1893, p. 86) there is a very singular mistake. He speaks of the
“cc ?
mantle-connection ‘ Bruecke ” between the two “ siphons,’ while he undoubtedly means the con-
’
nection in front of the branchial ‘ siphon.’ This is the one which is sometimes missing. The absence of the connection between the anal and branehial openings would be something unheard of; in fact, has never been observed by me in any member of the family.
® We must recall that similar variations or mutilations with regard to the mantle-connection which separates anal and supra-anal openings in the Unionid@ are well established, for instance in the genus
Fusconaia.
ORTMANN: SOUTH AMERICAN NAIADES. 463
it is positively known that no such mantle-connection exists, and the same is the case in all investigated species of Diplodon.
Thus it is seen that Castalina and Castalia (and possibly also Castaliella and Callonaia) form a group within the subfamily, which is not only specialized in the shape of the shell, but also in the tendency to close the branchial opening an- teriorly. As Castalina forms a transition from Diplodon to Castalia in the shell, so it is transitional also in this latter character, and thus it is perfectly clear that Castalina and Castalia cannot be primitive types, and other peculiar characters, which they possess, probably also should not be regarded as primitive.
This is especially true of the parallel ridges vertical to the edge of the lateral hinge-teeth (also sometimes present on the pseudocardinals). This character is best developed in Castalia (Plate XX XIX, figs. 8c, 8d), but Castaliella is also de- scribed as showing traces of it, and Callonaia has it. But here again Castalina is transitional toward Diplodon. Castalina nehringi has, according to Von Ihering, hinge-teeth obliquely corrugated, but never with parallel ridges vertical to the edge. This is quite correct, as far as concerns the laterals; but some of my specimens show that the furrows of the cardinal teeth occasionally may be nearly parallel, thus producing nearly parallel ridges. In Castalina martensi vertical ridges are present and well-developed, at least in the anterior section of the lateral teeth, but young specimens do not show them. My specimens of Castalina psam- moica have the vertical ridges poorly, or not at all, developed.
This sculpture of vertical ridges upon the hinge-teeth in Castalia and the allied genera has been much discussed in the literature. It is well known that Neumayr believed that these ridges are homologous to similar structures seen in the Trigoniidae, and that Castalia should be directly connected with Trigonia, and this idea has been taken up by other authors. However, Von Ihering (1893 p. 85) has pointed out that this is incorrect. From the above considerations we now know, that there undoubtedly is a genetic series, which leads from Diplodon through Castalina to Castalia, with the beginning at Diplodon, and the end at Castalia; and, consequently, the vertical ridges on the hinge-teeth in Castalia are not an ancestral character, inherited from Trigonia, but are a new acquisition, marking an independent, comparatively high development of the Naiad-hinge."°
Hyria and Prisodon may, or may not, possess these vertical ridges on the hinge- teeth. If present (Prisodon alatus, Pl. XL, figs. 2c, 2d), they are seen chiefly in
10 This idea has also been suggested by Pompeckj (Gegen Steinmann’s Geologische Grundlagen der Abstammungslehre, im 3ten Jahr.-Ber. Niederswchs. Geol. Ver., 1910, p. 12) in eriticizing Steinmann’s attempt at a polyphyletie derivation of the Naiades from various Trigoniide, Odhner (1918, p. 577,
footnote) also considers the sculpture of the teeth in Castalia as a secondarily developed character.
464 MEMOIRS OF THE CARNEGIE MUSEUM.
old specimens, while they are lacking in the younger ones, but old specimens also may not show them. Thus these two genera should be compared in this respect with Castalina, and it is quite possible that they have here their roots, the shape of the shell of Castalina also indicating in its elevated posterior slope the ‘‘alate”’ shape of Hyria.
Thus we see that the genera Castalina, Castalia, and probably also Castaliella and Callonaia, form a series descended from Diplodon, characterized by peculiar tendencies in the shell and in the soft parts. The shell tends to assume a more or less triangular shape, with high beaks, with a strong posterior ridge, and a trun- cated posterior slope. The hinge-teeth tend to develop vertical ridges, and the interdentum becomes wider. The soft parts tend to close the branchial opening in front. All the rest of the soft parts, however, remain here in a comparatively primitive state, being identical with, or standing very close to, the most primitive type of Diplodon in the interrupted character of the septa of the gills (See below). It also should be mentioned that the glochidia of these forms, where known, as in Castalina and Castalia, possess the same shape and the peculiar hooks observed in many species of Diplodon, but not in all (See below).
Apparently a side-branch of this series is formed by Hyria and Prisodon. Here the “‘alate’’ character of the shell, slightly indicated in Castalina, is empha- sized, and developed to an extreme degree; the hinge-teeth have a slight tendency to develop vertical ridges, exactly as in Castalina: but in the anatomy these forms do not show the tendency to close the branchial opening in front. Also in the rest of the soft parts they remain upon the primitive Diplodon-stage.
We may express the affinities as follows:
Callonaia Castaliella Prisodon Castalia Hyria Castalin Diplodon
The genus Diplodon yet remains to be discussed. We have referred to it above as the most primitive type within the subfamily, and there are undoubtedly a number of species contained in it which have a primitive structure. But there are others which differ from them. The first question, however, to be considered is which characters we should regard as primitive.
As far as the shell is concerned, there is considerable variety within this genus.
ORTMANN: SOUTH AMERICAN NATIADES. 465
Most of the species are subelliptical or subovate, and moderately elongated; but others approach more or less the suborbicular shape. For the latter, Simpson has introduced the subgenus Cyclomya, and it is true that these form a rather well- defined group, although there are certain species which seem to be intermediate.
The subquadrate or subtrigonal shape of the Castalia-group is unknown in Diplodon, and the well-defined posterior ridge characteristic of this group is very rarely found; the species, which have it, are always rather elongated (not trigonal), so there cannot be any mistake about them.
The hinge-teeth are also variable, but generally represent, or are easily com- pared with, the normal type described above (p. 459). There are one or two pseudocardinals and two laterals in the left valve, and two pseudocardinals and one lateral in the right valve. The pseudocardinals may be stumpy, or more or less compressed. They, as well as the laterals, may be smooth, or have corruga- tions, or may be split and dissected. They never have parallel ridges.
The beak-seulpture also varies a good deal. It is always radial, but the ridges composing it may be shorter or longer, finer or heavier, smooth, or dissected by growth-lines, granular, and sometimes they may be irregular.
But all of these characters of the shell are connected with each other by nu- merous transitions, and it is impossible to say that any one type is more primitive than another. It is also extremely difficult to arrange the species of Diplodon into groups according to these features. Simpson (1914 pp. 1225, 1228) has divided the genus into subgenera and subordinate groups, and has attempted to condense the results in the shape of a “key,” but this key is practically worthless, of which fact anyone may convince himself, when he tries to use it for the identification of a species.
However, when I studied the soft parts of the various-species at my disposal, I discovered that there are rather well-marked and apparently important c/raracters shown by the anatomy, especially by the structure of the marsupial part of the inner gill of the female, while the rest of the anatomy is the same in all species.
The interlaminar connections of the gills are extremely weak in the male of Diplodon, and in the outer gill of the female (which has the same structure as in the male). They correspond to the description given by me for Hyria and Castalia (Tetraplodon) (Ortmann, 1911a, pp. 115, 117). (See the figures of gills of various species on Plates XLV, XLVI, XLVI; also sections of the gills on Plates XLVIT and XLVIII). These interlaminar connections are few and scattered over the face of the gill, and do not form distinct septa (thus resembling the condition seen in the Margaritanide). But in the marsupial part of the inner gill of the female,
466 MEMOIRS OF THE CARNEGIE MUSEUM.
these connections become more frequent, are heavier, and stand closer together. Very often we see short connections grouped together either in a reticulate form, or in rows parallel to the gill-filaments, 7e., vertical to the edge of the gill (See text- figs. 1 and 27, on pp. 455-6), and it should be noticed that precisely the same ar- rangement is also seen in the Australian Hyridella (Ortmann, 1912, pp. 100-103, fig. 1). The connections of adjoining vertical rows may alternate in an irregular way, and may thus form irregular transverse or oblique rows. Towards the edge of the gill, and near the base, the connections are often somewhat elongated, stand closer together in the same row, and thus a more distinct arrangement into vertical septa is brought about, separating more distinct water-tubes (ovisaes), which, how- ever, are laterally connected with each other by the interruptions of the septa (See Pl. XLV,.figs. 1b, 2b, 3; Pl. XLVI, figs. 1, 2,3, 4, 6, 7a, 7b; Pl. XLVIL fig. 1).
There is no doubt that the arrangement described above is primitive, for it most closely approaches the condition seen in the male gill. The marsupial strue- ture is simply brought about by a more frequent and heavier development of the interlaminar connections of the male, which approach each other, and partly ar- range themselves in vertical rows.
A step in advance is observed when the vertical rows prevail over the reticulate arrangement, and extend over the whole marsupial portion of the gill. Such cases are found, and in them the interlaminar connections may be shorter or longer, but they always stand close together in each row, so that we have all through the marsupium the appearance of distinet septa, which, however, are perforated by holes, so that the water-tubes (ovisacs) communicate with each other; this is most evident in Diplodon piceus (Pl. XLVI, fig. 2).
Furthermore in a few species (D. decipiens, Pl. XLV, fig. 4b, and D. mogymirim, Pl. fig. 5c; see also Pl. XLVII, fig. 7, and Pl. XLVIII, figs. 2a, 2b), I have observed a further advance in this structure. Here the septa become solid, the interruptions are missing, and each septum runs from the base of the gill vertically towards the edge, exactly in the manner which we know to be the characteristic condition in the Unionide. Well-defined water-tubes, which do not communicate, are thus formed. The species, which show this structure, have in the marsupial part of the gill hardly any indications of interruptions of the septa. In the non-marsupial por- tions, as well as in the outer gill of the female, and both gills of the male, the usual structure of Diplodon is present, showing few and scattered connections. This reveals that the solid septa of these species of Diplodon represent the highest stage in the marsupial development in this genus, and that this feature should not be considered as homologous to the septa of the Unionide. It is analogous, and has been independently acquired.
ORTMANN: SOUTH AMERICAN NATADES. 467
There are other differences in the marsupium of the species of Diplodon, and these depend upon its location within the gill. Tn the Australian Hyridella nearly the whole inner gill is marsupial, only very small portions at the anterior and posterior ends remaining non-marsupial, a very common condition in the Unionide where the whole outer gill may be marsupial. In Diplodon we sometimes observe a similar arrangement (O. piceus, Plate XLVI, fig. 2), but generally a more con- siderable part at the anterior as well as at the posterior end of the gill is non- marsupial, so that the marsupium is restricted to the middle portion (about half) of the gill. The marsupial part may become still more reduced in size, and may be located not in the middle of the gill, but more toward the front, or toward the pos- terior end. In either case the posterior or anterior half of the gill is nearly or quite non-marsupial. It should be noted that I have observed this shifting of the marsupial part only in cases where the marsupial structure is comparatively primitive, with the interlaminar connections arranged in a reticulate way or as interrupted septa, but never in forms with solid septa, where the marsupial part always extends over a large seetion in about the middle of the gill. (Compare figures on Pl. XLV, XLVI, XLVII.)
All these differences described in the structure and location of the marsupium are constant within the species. In some forms, indeed, my material is rather scanty ; but in quite a number of other cases I have a sufficient number of individuals with soft parts, and I have invariably found that all females of the same form and from the same locality agree with each other, with the only qualification that in young females the marsupium is generally less extended, and occupies a smaller section of the gill (Compare Pl. XLVI, figs. 5b and 5c; Pl. XLVI, figs. 7a and 7b).
It should be emphasized that these marsupial differentiations are only found within the genus Diplodon. The other genera of which I have anatomical material stand generally upon the stages which I have described as the more primitive in Diplodon. Firstly the structure of the marsupium is of the reticulate type in the middle, with more or less development of interrupted septa towards the edges of the gill in Hyria (Ortmann, 1911a, p. 115) and Castalia undosa (ibid., p. 117), while in Castalina nehringi the structure showing perforated septa prevails. Cas- talia acuticosta seems to agree with C. wndosa, but my material consists of only two gravid, rather young specimens, in which the structure cannot be clearly seen on aceount of the mass of glochidia filling the marsupium. Secondly the location of the marsupium in Hyria and Castalia is in the middle of the gill (one-fourth at anterior end, less than that at posterior end, non-marsupial). In Castalina nehringi (Plate XLVII, fig. 2) the marsupium has moved a little more backward with about
468 MEMOIRS OF THE CARNEGIE MUSEUM.
the anterior half of the gill, and less than one-fourth of the posterior end non- marsupial.
Thus it is clear, that, while the Castalia-Hyria-group of genera in the shape of the shell and the hinge-teeth, and the Castalia-group in the conformation of the branchial openings represent a higher specialisation, in the rest of their anatomy, and chiefly in their marsupial characters, they all remain very close to the primitive Diplodon-type.
THE GLOCHIDIA. (See Text-figure 4.)
In conclusion I may say that I have found differences in the glochidia within the genus Diplodon. That certain South American species of ‘‘ Unio” have glochidia, was first announced by Lea in the case of U. peculiaris and U. firmus (Lea, Obs., XIT, 1869, Pl. 34, figs. 80, 82; first published in 1868). He describes and figures them as subtriangular in outline, oblique, or upright, the ventral margin with a point, and in the text he says that they are “furnished with hooks.’’ However, he neither describes nor figures these hooks.
Von Thering (1893, p. 47) states that hooks are missing in all of the South American species examined by him, but that the larvee are true glochidia.
This is about all we have known hitherto about the larval form of the South American ‘‘Unios” = Hyrtinw. I have now found that the normal shape (out- line) of the glochidia of Diplodon, and of the subfamily Hyriina, is as described by Lea, namely subtriangular, with a point on the lower margin (see fig. 4). In addi- tion, I have found that some species actually possess a “hook; but this hook is entirely different from the one so well known in the European genus Unio and in the Anodontine of Eurasia and North America. The very fact that Lea mentions the hook in a kind of perfunctory way, not calling attention to its peculiar features, suggests that he never saw the real hook, and that he simply took the point of the lower margin of the glochidium for it.
This Hyriine hook (Fig. 4, b, c, d, e, f, h, k, l, m) differs entirely from the Ano- dontine hook. The latter is triangular, attached by a broad base to the point of the lower margin, and carries upon its upper surface a number of fine spimules. The Hyriine hook is long and narrow, spiniform, with very narrow base, articulated to the point of the lower margin, and without any spinules on the upper face, and furthermore has a peculiar S-shaped curve.
It is perfectly clear that this hook is different from that found in the genus Unio and the Anodontine. Functionally it may serve the same purpose, that of
ORTMANN: SOUTH AMERICAN NAIADES. 469
forming attachment to fishes, but we have not the slightest direct evidence of this, and Von Thering directly questions it (J. ¢., p. 47). Morphologically this organ has been independently developed. It is analogous to the Anodontine hook, but not homologous.
Fie. 4. Glochidia of Hyriine, enlarged fifty times (drawn from photographs). Diplodon hasemani Ortmann (Rio Guaporé), from specimen No. 6. Cat. No. 61.5857. D. imitator Ortmann (Santa Maria), from specimen No. 22. Cat. No. 61.9248.
D. simillimus Ortmann (Morretes), from specimen No. 2. Cat. No. 61.9250.
Seo OS
D. vicarius Ortmann (Aqua Quente), from specimen No. 9. Cat. No. 61.9251 (shell of this specimen figured on Plate XXXVI, fig. 2).
e. D. decipiens Ortmann (Serrinha), from specimen No. 10. Cat. No. 61.9253.
f. D. paulista (Von Ihering) (Mogy Mirim), from specimen No. 10. Cat. No. 61.9256. g. D. charruanus (D’Orbigny) (Santa Isabel), from specimen No. 5. Cat. No. 61.5861. h. D. piceus (Lea) (Uruguayana), from specimen No. J. Cat. No. 61.5862.
i. D. hilde Ortmann (Cachoeira), from specimen No. ec. Cat. No. 61.5864.
k. D. mogymirim Ortmann (Mogy Mirim), from specimen No. 48. Cat. No. 61.9260.
l. Castalina nehringi Von Ihering (Salto das Cruzes), from specimen No. 1. Cat. No. 61.5119. m. Castalia acuticosta Hupé (Rio Guaporé), from specimen No. 1. Cat. No. 61.5112.
I haye said that only some Hyriinew have this hook. I found it in the case of several species of Diplodon, in Castalina nehringi, and Castalia acuticosta, and it
470 MEMOIRS OF THE CARNEGIE MUSEUM.
should be mentioned at this point that these hooks appear only in the fully developed glochidium, while in younger, immature glochidia they are lacking.
But it seems that in certain species of Diplodon the hooks are always absent, and in these cases the glochidia fully resemble the figures given by Lea. Of course, in certain cases it is hard to decide whether the glochidia are fully developed, or whether the absence of hooks may be due to the immature condition of the indi- vidual. However, I have a case in which I believe I am justified in thinking that the glochidia, when fully developed, have no hooks. In three gravid females of D. charruanus, the glochidia were all alike, and no hooks were observed (Fig. 49). One of these apparently was discharging, and thus we should expect mature glochi- dia (unless this were a case of premature discharge).
In a few other cases I am sure that mature glochidia have no hooks, since they are surrounded around the whole lower edge, with a margin, which possibly represents the first beginning of the permanent shell of the adult (Fig. 4a, 7). In other groups of Naiades, with one exception (Anodonta imbecillis Say of North America), nothing of the kind is known in glochidia, as long as they remain within the marsupium, and also in these South American forms, when immature, the glochidia do not possess this margin. I never have seen a trace of hooks here. The margin has much the appearance of that formed in young North American Unionide, after the parasitic stage on fish.
That in these cases the shell should appear at so early a stage, when the larva is still within the marsupium of the mother, is indeed remarkable, and possibly points to the conclusion that the modes and conditions of embryonic develop- ment in the Hyriine differ considerably from those of the Unionidae.
If the above observations are correct, we would have three types of glochidia in the genus Diplodon: (1) Of triangular shape, with hooks; (2) Of the same shape, without hooks; (3) Of the same shape, and without hooks, but with a margin around the lower edge, which obliterates the triangular shape.
Whether the second group is real, or only due to incomplete observation, or passes finally into the third, remains to be seen. At any rate, it is very desirable that close attention should be directed to this question in future work on South American Naiades.
The size of the glochidia is comparatively large, varying from 0.20 to 0.85 mm., which might be called a good medium size in comparison with the glochidia of the Unionide.
I regret that my observations on the glochidia of the Hyrvine are not more satisfactory. There are indications of important differences, but for the present
-
ORTMANN: SOUTH AMERICAN NAIADES. 471
we must be satisfied with having pointed out this fact, and with hinting that per- haps further knowledge of the glochidia may furnish at least some criteria for a classification of the species of Diplodon. The same remark holds good of the mar- supial structure. In reference to this we possess a little more knowledge, but it is still too scanty to make an attempt to use it for classification. The genera with the specialized shells of the Castalia- and Hyria-groups are well defined, but their anatomy and their glochidia do not present any remarkable differentiations, except the structure of the branchial opening in the Castalia-group. On the other hand, it appears that the genus Diplodon may finally prove to be composed of an aggrega- tion of more varied forms than heretofore supposed, and that the distinctive char- acters of these are found chiefly in the marsupium and the glochidium.
We may perhaps distinguish within the old genus Diplodon a type, which we might regard as having the more primitive features, such as a marsupium com- posed of interrupted or reticulated septa, occupying the whole or nearly the whole of the inner gill, and probably one of the types of glochidia above described. But it is hard to say, which form this is. I am inclined to regard the hooked glochidium as the more primitive form. The margined glochidium certainly represents a more advanced type. It might finally be possible to split up this genus into smaller genera, one of which should contain these primitive forms, the others to include forms more advanced in regard to their marsupial structure and glochidia. But in the present state of our knowledge this step cannot be taken, and we must be
satisfied with an arrangement of the species based upon the characters of the shells.
THE SPECIES OF THE Hyritn. Genus DreLopon Spix (1827).
Diplodon Srrx, 1827, p. 33 and plate 26.—DreLopon Simpson, 1900, p. 872; 1914, p. 1224.4 Type of genus:—Diplodon ellipticum (ellypticum err. typogr.) Sprx, 1827, PI. 26, figs. 1, 2. (Same type given by Simpson, 1900.)
THE SUBDIVISIONS OF THE GENUS DIPLODON.
Simpson (1914, p. 1225) has made an attempt to divide this genus into sub- genera and groups. But he also included in it Australian species under the sub- genus Hyridella Swainson, which we now regard as a separate genus, and an African
1 Diplodon, as accepted by H. & A. Adams (1858, p. 497) as a subgenus of Unio, is an entirely
heterogenous association of species, including forms from all over the world.
472 MEMOIRS OF THE CARNEGIE MUSEUM.
form (Subgenus Levirostris Simpson), which we may with considerable confidence assume does not belong here, until the contrary has been positively demonstrated.
Thus restricted, the genus Diplodon from our point of view falls in Simpson’s arrangement into two subgenera, Diplodon (Simpson, 1914, p. 1226) and Cyclomya (Simpson, 1914, p. 1278), each subdivided into groups. The characters of the shells of these groups run into each other very insensibly, and are found in various combinations. The subgenus Cyclomya is somewhat better defined, inasmuch as the rounded outlines of the species assigned to it contrast rather markedly with most of the elliptical, somewhat elongated shells, assigned by Simpson to typical Diplodon. But there are intergrades even here. Typical Cyclomya (of the fune- bralis-type) ts irregularly circular, with rounded angles (obscurely pentagonal), and has a short and high shell, with the greatest height situated in the middle of the shell, at about the middle of the ligament. Simpson included in it as species fontainianus, and gratus, which I regard only as higher and shorter forms belonging to Diplodon. They have the greatest height of the shell not in the middle, but more posteriorly, behind the ligament, and thus their outline is distinetly oblique and subtrapezoidal, although approaching the rounded shape.
Simpson, indeed, says in addition, that the beak-sculpture in typical Diplodon
b)
consists of “unbroken ridges,” and that in Cyclomya it is “irregularly radial.” In both cases this holds good only for certain species, and cannot be used as a
generally distinguishing character.
Subgenus DipLopon Simpson. Simpson, 1900, p. 873; 1914, p. 1226.
Shell more or less elongated; elliptical, ovate, or subtrapezoidal in outline; when short, more or less angular and distinctly oblique, with the greatest height behind the ligament, but not subcircular with the greatest height under the ligament.
The arrangement into groups, as here given, does not rest upon that of Simpson. It is largely made to suit my material, and does not claim to be final. We may ex- press the essential differences in the following key, but with the distinet under- standing that there are transitions between the groups, which are hard to place.
Key Tro Groups IN GeNusS DrIpPLopON.
a. Shell straight, not oblique, 7.e., the longest axis is nearly parallel to the ligament. b. Beak-sculpture covering a considerable part of the disk; ridges, chiefly the posterior, rather 2th a oe ae Ae OER MA aS Ok rc Gicopoeeas Subs od odo est rs Group of D. hyleus. bb. Beak-sculpture more or less developed, rarely covering a considerable part of the disk; ridges
not very heavy and rather uniform.
ORTMANN: SOUTH AMERICAN NAIADES. 473
c. Beak-sculpture well-developed, fine; ridges cut up into fine nodules...Group of D. granosus. ce. Beak-sculpture restricted to the region of the beaks; ridges fine, simple, not granular.
d. Shell rather compressed and flattened upon the sides, subtrapezoidal in outline, and
not distinctly pointed behind..........------+++++++ssrres: Group of D. chilensis. dd. Shell slightly or not at all compressed, convex upon the sides, mostly subelliptical or
subovate in outline, more or less pointed behind. e. Shell rather elongate, not high, subelliptical or subovate. . .Group of D. charruanus. ee. Shell shorter and higher, subovate or subtrapezoidal....... Group of D. lacteolus. aa. Shell distinetly oblique, with the longest axis forming an angle with the line of the ligament. Outline
subovate or subtrapezoidal, sometimes rather high and short........----- .Group of D. ellipticus.
1. Group or Diplodon hyleus.
Shell subelliptical, subovate, or subtrapezoidal, more or less pointed posteriorly, straight, not distinetly higher behind, nor oblique. Beak-sculpture well-developed, covering a considerable (but variable) part of the shell; bars rather heavy; those upon, and immediately in front of, the indistinct posterior ridge of the shell are heavier and longer than the rest; one or two of the median bars joining at their lower ends; sometimes the bars are somewhat nodulous (but not granular).
The beak-sculpture is the most essential feature of this group. Additional, but subordinate, characters are found in the hinge-teeth, chiefly the larger ones in each valve, which are triangular, rather thick, ragged, but not compressed. Often there are two pseudocardinals in the left valve.
I have seen the female soft parts of two species, D. hasemani and trifidus. In both the marsupial part of the inner gill is located in the middle section and has interrupted septa. The former species has the margined type of glochidium. It seems that this group is not very primitive.
1. DipLopon nyLaus (D’Orbigny). Unio hylwa D’OrBrany, 1835, p. 36; 1843, p. 607, Pl. 69, figs. 8, 9. Unio hyleus Sowrrsy, XVI, 1868, Pl. 93, figs. 506 a and 6 (poor figures). Margaron (Unio) hyleus Lea (pro parte), Syn. 1870, p. 31. Unio hyleus Von Martens, 1894, p. 164. Diplodon hyleus Simpson (pro parte), 1900, p. 884; 1914, p. 1274.
Type-locality.—‘ Rio Palometas, Rio Pari, and Rio Tucabaca, in the provinces of Santa Cruz de la Sierra and Chiquitos, in Bolivia.”
I have been unable to locate the first two rivers; a Rio Tucaraca, in the Chi- quitos country (Prov. Santa Cruz de la Sierra in Bolivia), runs to the Paraguay River, and I have no doubt that this river was intended, and it is advisable to take this as the type-locality.
474. MEMOIRS OF THE CARNEGIE MUSEUM.
Other localities.—Von Thering (1893, p. 119) gives this species from Rio Para- guay, but also upon D’Orbigny’s authority,” from the Amazon-drainage. Von Martens (J. c.) reports it from Paraguay. No other localities have been hitherto cited.
New locality—Rio Paraguay, Sao Luiz de Caceres, Matto Grosso, Brazil, (J. D. Haseman coll., May 25, 1909). One specimen.
Distribution.—The new locality is not very far from the type-locality (Rio Tacaraca), about 400 kilometers up the river, and demonstrates that this species belongs to the upper Paraguay drainage in Bolivia and western Brazil. According to Von Martens it goes down this river to Paraguay.
D. hyleus and guaranianus have been united by Lea, Sowerby, and Simpson, but I think that this is not correct, and that I have both species before me, agree- ing well with D’Orbigny’s remarks about them. The real D. hyleus is a larger shell, with the lower posterior end somewhat more produced, and with differences in the beak-sculpture, which extends over a larger section of the shell, and con- sists of rounded bars, which are somewhat irregular, and are rugose or slightly nodulous. The figure of D’Orbigny (Fig. 8) shows the sculpture very well.
I have only a single individual of this species, without soft parts, and have drawn the following description from this.
Description of Shell.—Shell comparatively small, moderately solid, subovate or subtrapezoidal, slightly higher behind. Height 64 pr. ct. of length. Dorsal margin very gently curved, passing into the posterior margin in a blunt, rounded angle. Posterior margin obliquely descending, emarginated, but this emargina- tion undoubtedly is an individual feature, since the growth-lines indicate that it was not present, when the shell was younger. Lower posterior angle slightly pro- duced, but rounded. Lower margin with its lowermost part placed at between two-thirds and three-fourths of the length of the shell (from anterior end), curving up behind. In the anterior portion it slopes upward in an almost straight line, finally curving up into the anterior margin; thus the shell appears slightly narrower in front than behind.
Valves rather flat (in this respect my specimen differs from the original hyleus, which is more convex), gently and rather uniformly convex, with the umbonal (posterior) ridge weakly marked, and indicated chiefly by a shallow radial de- pression upon the posterior slope, which forms the emargination at the posterior margin, and makes the posterior ridge appear slightly biangulate towards the
“ He says that all forms reported by D’Orbigny from Bolivia are from the Amazon-drainage. In
this particular case this is not correct, since the Rio Tucaraca of Bolivia drains into the Paraguay.
ORTMANN: SOUTH AMERICAN NAIADES. 475
posterior end. Diameter of shell 33 pr. ct. of length. Beaks not swollen, and not prominent, located at about one-fourth of the length from the anterior end.
Beak-sculpture strongly developed, covering about half of the disk, the longest bars (near the umbonal ridge) are about 20 mm. or more long. There are about sixteen or seventeen radial bars, of which the ninth and tenth, and the eighth and eleventh, unite in sharp angles, and between the ninth and tenth, there is a short odd bar, which is indistinct on account of the erosion of the beaks. Posteriorly the bars increase in length as well as thickness: the anterior bars are comparatively sharp, narrow, but much injured in my specimen; the posterior bars are broader and rounded. Upon the umbonal ridge, the radial bars are again finer, and upon the posterior slope there are four or five additional fine bars, which are shorter, and restricted to the upper part of the slope. On the lower part of the latter, there are a number of fine, irregular, oblique wrinkles. The lower ends of the radial bars are cut up, chiefly near the umbonal slope, into irregular, low tubercles, and traces of such tubercles may be seen near the lower margin of the shell. No distinct lunula is seen in our shell, but this part of the shell is badly eroded.
Epidermis with numerous, irregular, concentric wrinkles, and traces of radial lines. Color brown, much like that of D’Orbigny’s figure.
Hinge-line gently curved. Ligamental sinus over the posterior fourth (or a little more) of the lateral teeth, which are gently curved, one in the right, two in the left valve. Pseudocardinals directed obliquely forward and downward, two in right valve, the anterior one narrow, low, and compressed, the posterior one triangular, cut longitudinally into two parts. In the groove behind this tooth, the hinge-line has two small denticles. In the left valve, there are two pseudoeardinals, the anterior subtriangular, slightly compressed and simple, fitting into the groove between the two teeth of the right valve, the posterior one broader, and cut into three parts. Leaving out of sight the comparatively stumpy and double character of these teeth, their finer structure can only be regarded as an individual characteristic.
Cavity of shell and beaks shallow. Nacre whitish (discolored in the cavity). Anterior adductor-sear distinct, subelliptical; anterior retractor-scar separated from it, small, round and deep; anterior protractor-sear connected with the ad- ductor-sear. Posterior adductor-scar faint, subovate, with an upper triangular process formed by the posterior retractor-scar. Mantle-line faint. Dorsal muscle-sears in the cavity of the beaks.
476 MEMOIRS OF THE CARNEGIE MUSEUM.
MEASUREMENTS.
| Length. | Height. | Diameter | Beaks. My specimen............|42mm.| 27 mm. =64 pr. ct. of L. | 14 mm. =33 pr. ct. of L. | at 10 mm. =24 pr. ct. of L. D’Orbigny’s figure... .... ASiees 25. >) =58 a Ue ES Sa7/ af Se oa 19 ot Do., in text ie AO 60 <¢ 40
Remarks: The chief characters of this species, so far as I am able to make them out from the figure and description, and the single specimen at hand, by which it differs from the other species of this group, are found in the general shape (moderately elongated, slightly wider behind, with rounded posterior angle), the beak-sculpture, which covers a rather large portion of the disk, and the rough character of the posterior slope. The greater compression of our shell probably is individual, and also the emargination of the posterior margin.
2. DIPpLOoDON GUARANIANUS (D’Orbigny).
Unio guaraniana D’OrBtGNy, 1835, p. 37; 1843, p. 608, pl. 69, figs. 10-12.
Type-locality.—Rio Parana, Itaty, Province of Corrientes, Argentina. No other locality previously known.
New Localities—Paraguay River, Corumbi, Matto Grosso, Brazil (H. H. Smith coll.). Onespecimen, juv. In swamps of Lambaré, near R. Paraguay, Asun- cién, Paraguay (J. D. Haseman coll. March 31. 1909). One complete specimen, 6 left valves. Rio Paraguay, Sido Luis de Caceres, Matto Grosso, Brazil (J. D. Haseman coll. May 25, 1909). One male, with soft parts, and seven specimens, shells only.
Distribution.—Middle Parana above the junction with the Paraguay River in Argentina and Paraguay, and Paraguay River through Paraguay as far as Matto Grosso, Brazil.
I disagree with previous authors, and regard this species as being distinct from D. hyleus. My specimens answer very well to the description and figures of D’Orbigny, and they differ from hyleus in their somewhat smaller size, in the more distinetly truncated (more steeply descending) posterior margin, forming a more distinet lower posterior angle, but chiefly in the beak-sculpture, which consists of a smaller number of bars, which are heavier, chiefly posteriorly, and are not so rugose. My specimens are also more swollen than the single individual of hylaus at hand, but the latter is, as has been stated, probably exceptional in this.
The beak-sculpture is somewhat variable, chiefly in the length of the bars, which are from 15 to 20 mm. long, and generally cover more than half of the disk, but in the larger shells, sometimes less. There are twelve to fourteen radial bars,
ORTMANN: SOUTH AMERICAN NAIADES. 477
of which the seventh and eighth, or ninth and tenth, unite in the middle of the valve in a sharp angle; sometimes a short odd bar is found between these pairs. The bars are rather fine and sharp in front, but those behind, near the umbonal ridge become broad and rounded, their lower ends being irregular and indistinctly tubercular (much less so than in D. hyleus) occasioned by concentric lines cutting across them. These posterior bars also are distinctly longer than the anterior ones. Sometimes there are a few additional fine bars near the beaks on the posterior slope and below them a few oblique wrinkles.
The posterior end of the shell is also somewhat variable, but it never is rounded and slightly biangular, as it is in D. hyleus. The posterior margin is obliquely descending, and the lower posterior end is bluntly pointed, the angle being more or less prominent. The lower margin of the shell is evenly convex in some speci- mens, in others it is more strongly convex a little back of the middle, forming a blunt angle.
The specimen from Corumba is young, and the beak-sculpture covers all of the shell. It is also a little more compressed than the others, but agrees with them in all other characters.
MEASUREMENTS. Length. | Height. Diameter. Beaks.
Asuncion, No. 1......... 21 mm. 14 mm. =67 pr. ct. of L. 10 mm. =48 pr. ct. of L.. at 4. mm.=19 pr. ct. of L.
Do. Dit aretetayy «ts 27 18) - =67 a 14 SS =52 He 6.5“ =24 =
Do. hires ho ate ove 28 a 19 “ =68 ad 13 40) s 7 * =25
Do. Be tees 33.5 “ | 22 “ =66 4 13 “ =39 4 8 =24 ‘ S. buisids Ca(G))\.... o-2- 108 “| 24 “ =62 a 15.5 =40 8 =21 5 DiOrbienyis. fees woe. eloe h 18) 9 Se Sa) = 12 “) -=38 SF 8 =25 s
| !
Thus my specimens are on the average slightly higher and shorter, and some- what more swollen than the original, unless there are inaccuracies in the figure, which is not impossible, since the figure is 32 mm. long, while the text gives the length as 21 mm.
Anatomy.—The only specimen (8S. Luis de Caceres) of which soft parts are at hand, is a male, and has the structure typical of the genus. The anal open- ing is slit-like, and about three-fourths of the length of the branchial. The branchial opening has fine papille. The palpi are triangular and rather small, the posterior margins are not connected. Inner lamina of inner gills connected with abdominal sac. Gill-structure typical.
478 MEMOIRS OF THE CARNEGIE MUSEUM.
3. DIPLODON HASEMANI Ortmann, sp. nov.
Pl. XXXIV, figs. 1 to 4, shells; Pl. XLVII, fig. 5, section of gills; text-fig. 4a (p. 469), glochidium.
Locality.—Rio Guaporé, near Rio Sao Simao, Matto Grosso, Brazil (John D. Haseman coll., July 20, 1909). Nineteen specimens investigated, all with soft parts; sex of the three smallest not ascertained; the rest were males and gravid females.
Type-set: Carn. Mus. No. 61.5857, twelve specimens, among them four males and five gravid females, one with eggs, the others with glochidia.
Description of Shell.—Shell small (max. length 28 mm.), solid, swollen, short, subovate, somewhat pointed behind. Diameter 67 to 72 pr. et. of length, as against 59 to 68 pr. ct. in D. guaranianus. Valves not gaping. Dorsal margin gently convex, passing gradually or with a blunt angle into the anterior margin, which is sometimes almost truncated. The posterior upper margin forms an obtuse, rounded angle with the posterior margin, the latter descending obliquely, gently convex, forming with the lower margin a rather distinet, but blunt posterior ter- mination of the shell, which is little elevated above the base-line. Lower margin gently and uniformly convex, passing in a regular curve into the anterior margin. The anterior portion of the shell does not appear appreciably narrower than the posterior, or very little so.
Valves convex, more so in older specimens, with a rather distinet, but rounded umbonal ridge. Posterior slope subtruncated, a little compressed and elevated in the middle. Greatest diameter of shell 42 to 55 pr. ct. of length (82 to 52 pr. ct. in D. guaranianus). This greatest diameter located more forward toward the beaks than in D. guaranianus. Beaks somewhat swollen, but little elevated above the hinge-line, located at 25 to 28 pr. ct. of the length. Beak-sculpture of the hyleus-type, strongly developed, the posterior bars thicker and longer than the anterior. They cover 10 to 15 mm. of the shell, that is to say, hardly half of it in larger specimens. There are from fourteen to sixteen of them, and the ninth and tenth generally meet at an acute angle, with sometimes a short odd one between this pair. The anterior bars are sharp, the posterior ones broader and rounded, but not so much as in D. guaranianus. There are often three or four additional finer bars upon the posterior slope, and some oblique wrinkles behind and below them. The lower ends of the bars are but faintly cut up into tubercles. A lanceo- late, rather short, lunula may be present in the larger specimens.
Epidermis with numerous, concentric, irregular striz, finely lamellar in young
ORTMANN: SOUTH AMERICAN NAIADES. 479
specimens, and in older specimens on the posterior slope and toward the lower margin. Fine, obscure, radial lines all over the shell. Color rather uniformly dark brown, but in young specimens there is a distinct indication of a yellowish concentric band, or the region near the beaks may be of this color (suggesting the color-pattern of D. trifidus).
Hinge-line gently curved. Ligamental sinus over the middle of the lateral tooth, or a little farther back. Lateral teeth curved, one in right, two in left valve. In one specimen in the right valve there is a distinct, but low, accessory lateral below the normal one. Asa rule two pseudoecardinals in each valve. They are not much compressed and not much elongated, but rather stumpy, and are very much eut up, especially the posterior ones in each valve. The anterior pseudocardinal in the right valve is more distinctly compressed, but may be very small or even obsolete, and there may be a trace of a third pseudocardinal behind. The posterior pseudocardinal in the left valve is very variable in shape and size.
Cavity of shell and of beaks moderate. Nacre in all specimens whitish. Anterior adductor-scar deep, even in young specimens, irregularly rounded. An- terior retractor-scar separated from it, small and deep. Anterior protractor- sear united with the adductor-scar. Posterior adductor-sear less distinet, sub- triangular, with a short upper process formed by the posterior retractor-scar. Pallial line distinct. Dorsal sears few, located in beak-ecavity and close to hinge-
plate. MEASUREMENTS.
No. \Sex.| Length. | Height. Diameter. | Beaks. Figures.
EDs lieteapeleeo) mm.| 9 mm. =72 pr. ct. of L.) 5.5 mm. =44 pr. ct. of L.| at 3.5 mm. =28 pr. et. of L.
> aera be fan fet s 10 87, 7 “S47 Us + Ss =D. m
2a..| o | 18 Soe ides ati s 7.5 ‘ =42 Ey 5 “ =28 oe }
4.2 .| 2) 23 nS 16 i) 5S ll ee ees i Gia. = 28 “ | Pl. XEXEXIV, fie. 3. GES a ebay Se Maly Oe a 12 aff nf 7 aii, a
TeAolpeOus es 18 ‘ =69 es We OO Sie 2 :
9...) 9 | 28 | DEO C0) H 14 Eo = G0) 7 2.) " PPO RCV, figs, 2. 10...!o" | 27.5“ [20 ie Ge 15 =ab 7 = 26 = Pl. XXXIV, fig..1.
I cannot discover any sexual differences in the shell.
Remarks.—There is no question that this species is very closely allied to D. guaranianus, and it may be described as a small, short, rather swollen D. guarani- anus, with the posterior end of the shell a little sharper, the posterior ridge a little more distinct, and beak-sculpture less developed. Since I possess a good number of specimens of D. hasemani, and also a fair number of D. guaranianus, 1 do not doubt the specific distinctness, since the geographical distribution also differs. D. guaranianus belongs to the Parand and Paraguay Rivers, while D. hasemani is from the Guaporé, tributary to the Amazons-system. However, a former con- nection of these systems is indicated by the close affinity of the two species.
480 MEMOIRS OF THE CARNEGIE MUSEUM.
Anatomy.—I have before me many males and gravid females with soft parts. It should be noted that the smallest gravid females are only 16 mm. long and that this fact indicates that this is a small species, not growing to a larger size, as my material shows, reaching the maximum length of 28 mm.
Anal opening short, slightly shorter than the branchial opening, the latter . very short, with few but distinct papillae. Palpi subtriangular, lower margins convex, posterior margins searcely connected. Gillsnormal. Inner lamina of inner gill entirely connected with abdominal sac. In the larger gravid females the middle of the inner gill is marsupial, leaving a little less than one-fourth of the gill both at the anterior and posterior ends non-marsupial. In the smaller females the marsupial part is smaller. The swelling of the charged marsupium is moderate and the gills are charged to near the edge. In the marsupial part the interlaminar connections are arranged in interrupted vertical septa. Since no sterile females are at hand, the exact arrangement could not be seen in a face view, but from sec- tions (Plate XLVII, fig. 5) it is evident that it is very likely more or less reticulate.
Glochidia (Text-fig. 4a, p. 469) subtriangular and margined, without hooks. Measurements, without margin: L. 0.30 to 0.31 mm.; H. 0.24 to 0.25 mm.; with margin: L. 0.35 to 0.86 mm., H. 0.29 to 0.30 mm. The presence of eggs and glochidia on July 20 should be recorded with respect to the breeding season.
Color of Soft Parts in Alcohol.—Foot blackish in its distal part, this color separated in a sharp line from the light basal part. Rest of soft parts whitish.
4. DiIpLopon TRIFIDUS (Lea) (1860). Diagram of Soft Parts (See Text-fig. 1, p. 455). Unio trifidus Lma, Obs. X, 1868, Pl. 44, fig. 295. Diplodon trifidus Simpson, 1900, p. 884; 1914, p. 1272.
Type-locality—‘‘ Buenos Ayres.”’ Never reported again since its original description.
New Localities —Centre of Rio Guaporé, near Rio Sao Simao, Matto Grosso, Brazil (J. D. Haseman coll., July 20. 1909). Six specimens, all with soft parts. In the three smallest the sex is uncertain (they have the male structure); the others are two males and one (the largest) a gravid female with eggs. Rio Guaporé, Sao Antonio de Guaporé, Matto Grosso, Brazil (J. D. Haseman coll., July 31, 1909). One male with soft parts.
Distribution.—I have doubts as to the correctness of the loeality originally given. Von Ihering (1893, p. 118) lists this species with those from the La Plata drainage, but he rests entirely upon Lea. The latter received his single specimen
ORTMANN: SOUTH AMERICAN NAIADES. 481
from D’Orbigny with the above locality; but it is quite possible that a mistake was made, since D’Orbigny also collected in the Amazon system in Bolivia, in the general region, whence our specimens come. Nobody else ever re-discovered this species in the La Plata, although frequent collections have been made. Our material is beyond any doubt this species, and the locality is authentic.
The set, although containing only six specimens, comprises young and old, males and females, and thus it is worth while to give a full description.
Description of the Shell—Of medium size, growing larger than any of the preceding species (maximum length 58 mm.; the type is 42 mm., according to Simpson), rather solid. Outline subelongated, subelliptical, or long-ovate, rounded in front, pointed behind. Height 41 to 54 pr. ct. of length. Valves not gaping. Dorsal margin straight in young specimens, very gently curved in older ones. Anteriorly the dorsal margin forms an indistinct angle with the anterior margin in young individuals; in old ones it passes into it gradually. Posteriorly the dorsal margin forms a very obtuse angle with the posterior margin, or, in the largest specimen, passes gradually into it in a gentle curve. Posterior margin obliquely descending, gently curved, meeting the posterior portion of the lower margin in a blunt, but distinet, point, which is elevated above the base-line, but nearer to the latter than to the line of the upper margin. Ventral margin gently and rather regularly convex, its lowest point slightly behind the middle of the shell, ascending in front and behind it. In front it curves up into the anterior margin. Thus the shell has a long-ovate, almost subelliptical outline, with the anterior end only slightly narrowed and rounded, and the posterior more tapering and bluntly pointed.
Valves moderately convex, convexity rather uniform all over the disk, but strongest near the beaks and upon the umbonal ridge, which forms a rounded, but rather distinct, angulation running toward the posterior point. Posterior slope somewhat compressed. Diameter 23 to 36 pr. ct. of length. Beaks not swollen, hardly elevated above the hinge-line, located at 18 to 25 pr. ct. of the length. Since the shell is more swollen towards the beaks, the latter appear depressed, chiefly in old specimens. Beak-sculpture distinct and well developed, consisting of about sixteen radial bars; the anterior bars are shorter (about 5 mm. long), the posterior much longer, chiefly upon the umbonal ridge, where they are 15 mm. long, or even more. There are two systems of these bars; eight or nine anterior, and seven or eight posterior bars. In the middle of the shell one or two pairs unite at a sharp angle, and sometimes a short, odd bar stands between the innermost pair. The anterior bars are sharp; the posterior ones are also rather sharp near the beaks, but towards their lower ends they gradually become thicker and rounded, broaden-
482 MEMOIRS OF THE CARNEGIE MUSEUM.
ing and flattening, and finally disappearing. These broad bars are conspicuous chiefly immediately in front of the umbonal ridge, and are distinetly seen even in our largest specimens, where the beaks are greatly eroded. All these bars are smooth, except for the concentric strive of the epidermis which cut across them. In some of my young specimens there are a few oblique wrinkles upon the posterior slope, but there are hardly any radial bars, except one, fine, and short, close behind the broad bars of the umbonal ridge. There is a narrow and short lunula in front of the beaks.
Epidermis smooth, but with numerous, irregular, concentric lines, which be- come sublamellar upon the posterior slope and near the lower margin. There are also numerous fine, irregular, but straight, radiating lines on the disk below the beak-seulpture. The posterior slope has no radial ridges or furrows. Color of epidermis dark to light green, with concentric bands of yellow-brown. The three young specimens are light golden-brown towards the beaks, and light green towards the margins; in the old specimens the green color prevails, and becomes quite dark, but is interrupted by one or two lighter bands of brownish. No traces of color- rays present.
Hinge-line almost straight or very gently curved. Ligamental sinus over the posterior third of the lateral teeth. One lateral in the right, two in the left valve, rather long, distinct, and in old shells gently curving downward behind. Pseudo- cardinals very variable. Lea describes them as trifid in both valves, but this is not always the case. Of our largest specimen it may be said that the right valve has three teeth: the middle one the largest, directed obliquely downward and for- ward, triangular, narrow above, broader below, and deeply longitudinally cleft into three ridges; in front of it is an anterior, narrow, compressed tooth, which is connected with the middle one above; behind the largest pseudocardinal is a deep groove, followed by small, ragged elevation of the hinge-plate representing the third tooth. The left valve has one large posterior tooth, which is ragged and fits into the groove behind the middle tooth of the other valve (including the small elevation behind it), and in front of it is a narrow lamellar tooth, fitting into the space between the first and second teeth of the right valve. In the groove be- tween these two teeth of the left valve are two low ridges, corresponding to the clefts of the large tooth of the right valve. In our second largest specimen (No. 1) the same general arrangement is seen, but the anterior tooth of the right valve is very low, while the third (posterior) is more distinct and triangularly elevated. The left valve has two teeth. All these teeth are much less ragged, and the clefts of the middle tooth of the right valve are lacking, and also the corresponding ac-
ORTMANN: SOUTH AMERICAN NAIADES. 183
cessory ridges of the left valve. Similar, but rather variable, conditions are seen in the younger individuals and the specimen from $8. Antonio. The rule is that there are three teeth in the right, but only two in the left valve.
Cavity of the shell moderate, that of the beaks shallow. Nacre shining, snow- white in my specimens, in two (No. 1 and the one from 8. Antonio) with a faint salmon blush in the cavity of the shell. Anterior adductor-scar deeply impressed even in young specimens, irregularly rounded or broadly subelliptical. Anterior retractor-sear located above it, separated from it, deep, small. Anterior protractor- sear connected with it. Posterior adductor-scar distinet, but much less impressed, subovate or subtriangular, with a rounded or triangular appendix above, formed by the posterior retractor-scar. Pallial impression distinct. Dorsal scars five or Six, In cavity of beaks, placed irregularly, or in an oblique line.
There are no sexual differences in the shell.
MEASUREMENTS (SPECIMENS FROM SAO SIMAO).
No. Sex. Length. | Height. Diameter. Beaks. Choa anb6e ? 21.5 mm. | 9 mm. =41 pr. et. of L. 5 mm. =23 pr. et. of L. at 4.5 mm. =21 pr. et. of L. Bia oP rol 34 or 16 “ =47 Ni 10 “ =29 = 7 “ =21 ig ey me ter rot 56 A 30 “ =54 i 17 =3}1) 14 =25 CAS Ping SRa, 2 58 : 29 “ =50 a: 21 =36 14 =24 Lea’s figure... ... ZOE Ea If ey ‘! (i =36 7.5 =18
In Lea’s text there is apparently an error with respect to the height, since the measurement would yield the absurd figure of 89 pr. et. of the length. According to the figure Lea had a specimen with the beaks a little more anterior than any of mine, but otherwise rather closely agreeing with them.
Remarks.—This species is related to those of the hyleus-group, chiefly to D. guaranianus, but it is easily distinguished by its long-ovate shape, with pointed posterior end, and by the beak-sculpture, which, although of the same type, does not extend so far upon the disk. A good character is furnished also by the color of the epidermis and its concentric bands, though we have seen that this pattern is at least indicated in young D. hasemani. Lea compares D. trifidus with D. burroughianus and parallelopipedon, but there hardly is any close relationship to the former, and but a superficial resemblance to the latter. I think it belongs to the hyleus-group, and may be characterized as an extremely elongated member of that group, in which the beak-sculpture is not quite so fully developed as in the other species.
Anatomy (Text-fig. 1, p. 455).—The three smallest of my specimens could not be satisfactorily examined to ascertain the sex, but the gills did not show the mar-
484 MEMOIRS OF THE CARNEGIE MUSEUM.
supial structure, so that they may be males. Of the others, three are males, and one is a gravid female with eggs. The fact that this was collected on July 20 gives an indication of the breeding season.
Anal opening (a in text-figure) closed above, without forming a supra-anal. Closed part (s) two to three times as long as the anal; the latter (a) slit-like, short, shorter than the branchial opening. Inner edge of anal practically smooth. Anal separated from the branchial by a solid bridge (¢) formed by the union of the mantle-margins. Branchial opening (b) a little over twice as long as the anal, with distinct papillee on inner edge; mantle-edges not united in front of it. Palpi (h) subtriangular, the lower margins slightly convex, the posterior margins con- nected at the base.
Gills (¢ and 0) rather long and moderately wide. The inner (7) wider than the outer (0) chiefly in front. Outer gill narrowing behind and before, its anterior end near the highest point of the mantle-attachment-line. The inner gill has an almost straight lower margin, and is only little narrower anteriorly; its anterior end is immediately behind the palpi. Inner lamina of inner gill entirely connected with abdominal sac.
Structure of gills normal. In the male both gills have fine, scattered, and interrupted interlaminar connections, running parallel with the gill-filaments, but without forming complete septa or water-tubes. In the gravid female, the eggs are contained in the large middle section of the inner gill (7), leaving free less than one-fourth of the gill at the anterior and at the posterior extremity, and also leaving free a narrow zone along the margin. This marsupial part has the interlaminar connections strongly developed, in the shape of interrupted septa, forming incomplete, intercommunicating water-tubes, filled by the eggs in a dense mass, not separated into placentze. Since no sterile females are at hand, the exact arrangement of the interlaminar connections in a face view could not be made out. The charged marsupium is somewhat swollen and distended, so that the inter- laminar connections have stretched out. The outer gill of the female has the structure of the gills of the male.
2. Group or Diplodon granosus.
Shell subovate or subtrapezoidal, rather elongated, not distinctly pointed behind, straight, not distinctly higher in the posterior part, nor oblique. Beak- sculpture well developed, but fine, and characteristically cut up into numerous fine nodules or granulations, thus obscuring the radial arrangement. The granular sculpture often continued a good distance upon the disk.
ORTMANN: SOUTH AMERICAN NATIADES. 485
The essential character of this group is in the beak-sculpture, which, however, is very variable, although the granular structure is always more or less evident. It is hardly possible to give any additional characters, except that the hinge- teeth are moderately developed, subeompressed, and hardly ever stumpy. Two pseudocardinals are in each valve, but sometimes there are reductions.
I have very little material representing this group. The soft parts of only two specimens are at hand, one too young to be of any value, the other a male. Thus nothing can be said about the structure of the female. All my specimens seem to belong to one species, which is very variable, and has a very limited range in the coastal streams of eastern Brazil.
The genetic connections of this group likewise cannot be properly ascertained. The beak-sculpture does not seem to be very primitive, when compared with the other species of the genus. In the shape of the shell and the hinge, there is simi- larity to the next group (chilensis), which possibly may be the most primitive of the genus.
5. DreLoDON GRANOSsUS (Bruguiére) (1792).
Compare: Unio multistriatus Lea, Von InerinG, 1890, p. 165. Unio psammactinus BRONN, Von IHERING, 1893, p. 107. Diplodon expansus VON THERING, 1910, p. 134.
Diplodon granosus (BRUGIDRE), Stupson, 1914, p. 1250.
I am inclined to accept the earlier opinion of Von Ihering (1890), and that of Simpson, that a number of so-called “species,” of which Simpson has given a full synonymy, belong here. Possibly several others should be included.
Type-locality.— Brazil and Guiana.
==
Additional Localities.—Rio de Janeiro (psammactinus, Philippi) (Von Ihering) ; coastal streams in eastern Brazil, between Rio de Janeiro and Bahia, but not to the south of Rio (Von Ihering); Rio Negro, Prov. Rio de Janeiro, Distr. St. Rita (Dunker’s coriaceus, Von Ihering); Tayuara, Distr. Canta Gallo, Prov. Rio de Janeiro (Von Ihering); Rio Parahyba do Sul, Rio de Janeiro (Von Ihering); Nova Friburgo, Rio de Janeiro (Von Thering). Rio Paraguassi, Bahia (Von Ihering, 1910, p. 139, as multistriatus Lea). Rio Doce, Espirito Santo (Von Ihering, 1910, p. 134, as expansus Kuester).
New Localities—Mountain creek, Raiz da Serra, near Santos, Sio Paulo, Brazil (J. D. Haseman coll., July 26. 1908). One specimen with soft parts, male. Rio Ribeira, Iporanga, Sio Paulo, Brazil (J. D. Haseman coll., December 1. 1908). Two specimens, young, one with soft parts of the male type.
486 MEMOIRS OF THE CARNEGIE MUSEUM.
In addition, the Carnegie Museum possesses two specimens labeled “ellipticus, Brazil,” from the Holland Collection, and one specimen labeled ‘“multistriatus, Brazil,” from the Juny Collection.
Distribution —The new localities represented in the collection made by Hase- man extend the range southward (southwestward) along the coast of Brazil beyond Rio de Janeiro into the southern portion of the state of Sao Paulo in the small coastal streams. The established range thus reaches from the Rio Paraguassu in the North to the Rio Ribeira in the South.
Remarks.—My material is entirely insufficient for the study of the various forms regarded as belonging here. Originally Von Ihering was inclined to unite all these forms with granular beak-seulpture into one species; but later he divided them according to the character of the beak-sculpture, although he admits that there is great variation in this respect. Simpson unites them again, at least in part, but he lets granuliferus Dunker stand, and even adds a new species, D. semigranosus."
My specimens also vary in the development of the beak-sculpture and in the shape of the shell, but they might very well be forms of one and the same species. The two soft parts at hand are those of males, and the structure of the female is unknown.
The two young specimens from Iporanga show the granular beak-sculpture very well, and I should eall them D. granosus by all means. Simpson (1914, p. 1249) has described from the same river-system (Rio Ribeira) at Iguapé, Sao Paulo, Brazil (near the mouth), a D. mimus, and Marshall (1917, p. 383, pl. 51, figs. 3-6) has redeseribed and figured it. It is founded upon two specimens, larger than mine, and differing somewhat from each other in their proportions, and also from my specimens, but the differences may be individual. The measurements are:
Length. Height. | Diameter.
SHON Gooinagashosoherl| qh Won, 28 mm. = 76pr.ct.of L. | 17 mm. = 46pr. ct. of L.
WAT HS) OVI Miles Greteuctekss Gan-crn gi A b-Beea)y co | @27 160 ie | 15 «= 3:
My specimen............| 29 “ | 17.5 “ =60 «“ I Sgs5 oe eeeog a ener mene Soe & ee se
My specimens agree very well with the figures and description, except that there is no lurid-purple in the nacre, which is dull (lurid) whitish.
13 Habitats: “Sao Paulo River” (location unknown); “ Ponte Grande’ (location unknown); “ Qs Perus,” Sao Paulo, Brazil. Marshall (1917, p. 387) redeseribes and figures this species, and gives as type-locality: Rio Tieté, Sao Paulo, and the additional localities ‘‘ Ponte Grande, Sao Paulo; Os Perus: and Ponta Grossa, Parand.’’ The location of Ponte Grande is still unknown. Ponta Grossa
is on the headwaters of the Rio Tibagy, tributary to the Parand Panema.
ORTMANN: SOUTH AMERICAN NAIADES. 487
The beak-sculpture of Simpson’s species is unknown, being entirely eroded. It may be that my specimens are this, but I hesitate to unite D. mimus with granosus, before the beak-sculpture of the former has been described.
3. Group oF Diplodon chilensis.
Shell rather compressed, flattened upon the sides, subelliptical, subovate, subtrapezoidal, more or less elongate, not distinctly pointed behind, straight, and not distinctly higher in the posterior part, nor oblique. Beak-sculpture simple, with narrow, straight, uninterrupted radial bars, restricted to the region of the beaks, and not extending far upon the disk. Two of the median bars may be joined at their lower ends.
The generally subtrapezoidal shape of the shell, with flat sides, and the simple character of the beak-sculpture are the chief characters of the group, which may represent the most primitive type within the genus. The posterior end of the shell is mostly not pointed, but more or less rounded, and the posterior ridge is rather indistinct, not prominently angular.
A great number of ‘species’? belong here, which are extremely hard to dis- tinguish. It is in this group that I encountered the greatest difficulties in the identification of the species, and even at present I am not satisfied with the results. This is the more to be regretted, since I have abundant material with soft parts of some of these forms and have found that there are differences in the anatomy, which to all appearance are of specific value.
After many attempts to group these forms, I have finally concluded that the best way, the one that is least liable to lead into error, is to treat these forms geo- graphically. It is not very likely that the same identical species occurs in widely separated river-systems.
Forms of this type are found over nearly the whole continent, but apparently they are rare or missing in the region of the great depression which runs from the La Plata up the Paraguay to the Amazon-drainage. I shall begin with the forms from Chile and Patagonia, and proceed then in the direction toward Brazil, going from South to North.
1. SPECIES FROM CHILE AND PATAGONIA.
I have very insufficient material of this group, and cannot express any positive opinion about the species. But it seems that all, or nearly all, of the forms known from Chile belong here. Simpson (1914, p. 1257) admits ten of them: chilensis Gray, solidulus Philippi, gassiesi Kuester, aplatus Reeve, moline Philippi, modestus
488 MEMOIRS OF THE CARNEGIE MUSEUM.
Kuester, atratus Sowerby, obtusus D’Orbigny, chiloénsis Kuester, awreus Simpson. But the differential characters of these are very obscure, and I should not be as- tonished if some of these names should prove to be synonyms.
Similar forms are known from the eastern foot of the Cordilleras in Patagonia, in the drainage of the Rio Negro. Of these the following material is at hand:
6. DreLopon PpaTAGonicus (D’Orbigny) (1835).
Unio patagonicus D’OrBIGNyY, 1848, p. 610, Pl. 70, figs. 1-4. Diplodon patagonicus Simpson, 1900, p. 885; 1914, p. 1275; Prussry, 1911, p. 610.
Type-locality.—Rio Negro, 10-12 miles above its mouth.
New Locality.—Rio Limay, Patagonia (Received in exchange from W. Israél), one right valve.
Distribution.—Rio Negro and its tributary Rio Limay in Patagonia.
The specimen at hand is very poor, with the epidermis worn off, but it is un- doubtedly this species, which is characterized by its elongated outline and a shallow radial groove upon the posterior slope.
This species is not at all related to D. parallelopipedon (Lea), with which it has been associated by Simpson. It has, indeed, a similar elongated outline, but the characteristic strong and elevated posterior ridge of the latter species is entirely absent.
7. DIPLODON FRENZELI (Von Ihering) (1893).
Unio frenzelii Von InERING, 1898, p. 111, Pl. 4, fig. 12. Diplodon huapensis Bartscu, 1906, p. 394, Pl. 27, fig. 1; Pl. 28, fig. 1; Pl. 29, fig. 2. Diplodon frenzeli and huapensis, Simpson, 1914, pp. 1264, 1265.
Type-locality.— Patagonia.
Other Localities.—Patagonia, foot of the Cordilleras (Von Ihering); small lake on Victoria Island in Lake Nahuel Huapi, Argentina (Bartsch, hwapensis).
Locality Represented in Carnegie Museum.—Lake Nahuel Huapi, Argentina (C. H. Eigenmann coll., 1919). One gravid female with soft parts.
Distribution.— known from Patagonia, at the foot of the Cordilleras in the region of Lake Nahuel Huapi (Rio Negro drainage). The locality “Chile” given by Von Ihering is rather vague; and ‘Os Perus, Sao Paulo, Brazil,” given by Simpson for frenzeli is probably incorrect.
My material is too poor to give a full account of this species, but I am sure that my only specimen belongs here, and I also believe that huapensis is the same.
4 Unio patagonicus Reeve, XVI, 1865, Pl. 21, fig. 93, is not this.
ORTMANN: SOUTH AMERICAN NAIADES. 489
This species has the indifferent and uncharacteristic outline of the forms of the chilensis-group: subelliptical or subtrapezoidal, with upper and lower margins subparallel, and rather elongated and compressed shell. _ It is, however, remarkable for the anterior location of the beaks (18 or 19 pr. ct. of the length). There is no radial furrow or groove on the posterior slope.
The color of the epidermis of my specimen (the largest known) is dark brown, and the epidermis is wrinkled with concentric lamellae, but it has been largely eroded. According to Von Thering, the color is dark brown in the larger specimens, but dark olive in smaller ones, sometimes with lighter green in places. The color of huapensis is brown posteriorly, grading to wax-yellow anteriorly.
In all other respects, chiefly in the hinge-teeth, my specimen agrees with the description of frenzeli; only the posterior end of the shell is a little more broadly rounded, but not very different from the specimen figured by Von Ihering on plate 4, fig. 127.
D. huapensis has been compared by Bartsch with frenzeli, and he says that it can readily be distinguished from it by the narrower outline, that means to say by the height being less in proportion. This, however, is not correct, as can be seen by comparing the measurements.
The only difference I can see in huapensis is the more tapering posterior end. But since this species is founded upon two specimens, of which only one has been figured, this might very well be an individual character.
In the following measurements I leave out Von Ihering’s specimens from Chile, which appear to me a little doubtful.
MEASUREMENTS. Length. Height. Diameter. Beaks. Huapensis (figure).......... 53 mm. | 25.5 mm. =48 pr. ct. of L.. 14 mm. =26 pr. ct. of L. at 10 mm. =19 pr. ct. of L. Do. (type, text)...... }55 “ 25.9)! =A, ¥ 14.5 =26 os Do. ext) iiss oscs § Ui os Zico | = AS Sf 16 ‘ =28 Frenzeli 5b Von Ihering. ....| 56“ | 26 ee Se Gh 16 =29 “ LO Ls Do. 5a Von Ihering.....)58 ‘ | 28 “ =48 es 17 “ =29 s 11 “ =19 Do. Von Ihering........ 68 “ | 34 el si 20 0 . ie My specimen (9).......... Les | ater Siu! a 16 i 5} e 13 “ =18
Thus my specimen is more compressed than any of the others ; however, a variation of the diameter from 23 to 29 pr. ct. is not at all unusual.
Anatomy.—The specimen at hand has been preserved with the soft parts, and proved to be a gravid female with glochidia.
Color of soft parts whitish, with black pigment near anal and branchial open- ings extending forward a good distance along the margin of the mantle, and be- coming brown.
490 MEMOIRS OF THE CARNEGIE MUSEUM.
Bartsch has given a description of the soft parts of hwapensis, which refers chiefly to the general features, the color, the structure of the siphons, and the shape of the gills and palpi. But no particulars as to the structure of the gills are given.
Anal opening slit-like, closed above, closed part about four times as long as the open, the latter with the inner edge smooth, and separated from the branchial by a solid mantle-connection. Branchial opening with distinet papille. Palpi of moderate size, subtriangular, lower margins curved, posterior margins not connected at base.
Gills nearly of the same width posteriorly, the inner much wider than the outer anteriorly. Outer gill subtriangular, its anterior end at the highest point of the mantle-attachment-line. Inner gill with gently convex margin, very little narrower anteriorly, broadly attached in front, its anterior end immediately be- hind the palpi. Inner lamina of inner gill entirely connected with abdominal sac. Interlaminar connections of non-marsupial gills weakly developed, scattered. Marsupium located in the inner gill, but not occupying all of it, leaving nearly one-fifth of the gill free anteriorly, and about two-fifths posteriorly, so that the marsupium is distinetly shifted forward, lying in the second and third fifth of the gill (from the anterior end). Interlaminar connections of the marsupium forming interrupted septa, the septiform structure apparently prevailing (this is not quite clear, since the structure is obscured by the masses of glochidia). The charged marsupium is a little swollen, and the glochidia fill it in a loose mass, not being conglutinated.
Glochidium subtriangular, strongly oblique, with the point of the lower margin located vertically below the posterior end of the hinge-line. Hooks are present and of the normal S-shape. LL. 0.25, H. 0.20; L. of hooks: 0.05 mm. Compared with other species, the glochidium is rather small.
2. SPECIES FROM COASTAL STREAMS OF SOUTHERN BRAZIL.
It is a noteworthy fact that species of the chilensis-type seem to be absent from the system of the Rio de la Plata (with the exception of the drainage of the upper Parand). They also seem to be absent in the great Amazons-basin, and northward. But they are found rather plentifully on the Brazilian plateau, be- ginning at its southern extremity, in the coastal streams in Rio Grande do Sul, going thence northward, and crossing over in Sao Paulo and Paranda into the drain- age of the upper Parana River.
We shall take up first the species of the coastal streams, and I wish to call
ORTMANN: SOUTH AMERICAN NAIADES. 491
attention (as has been done by Simpson in the case of D. mimus) to the great re- semblance of these shells, not only among themselves, but also to the North American Elliptio complanatus (Dillwyn), distributed over the Atlantic streams of the eastern coast of the United States from Georgia to Maine. This resemblance, of course, is only external and superficial. Closer examination of the hinge, of the adductor-sears, and, if visible, of the beak-sculpture, at once reveals important differences. The anatomy is entirely different.
8. DreLopon IMITATOR Ortmann, sp. nov. Shells: Pl. XXXIV, figs. 5, 6,7; Pl. XXXV, figs. 1,2. Anatomy of gills: Pl. XLV, fig. 1. Section of gills: Pl. XLVII, fig. 6. Glochidiwm: text-fig. 4b (p. 469).
Type-locality—Rio Vaceahy-mirim, Santa Maria, Rio Grande do Sul, Brazil (J. D. Haseman coll., January 29. 1909). Twenty-three specimens, all with soft parts, among them males, barren and gravid females, with eggs and with glochidia. Type-set: Carn. Mus. Cat. No. 61.9248. (This is in the drainage of the Rio Guahyba-Jacuhy, far up in the headwaters).
Additional Locality.—Rio Jacuhy, Cachoeira, Rio Grande do Sul, Brazil (J. D. Haseman coll., January 26. 1909). One barren and one gravid female, with eggs, both young. (This is farther down, in the middle part of the Rio Jacuhy.)
(Originally there were thirty-four specimens from the type-locality at my disposal, all with soft parts.)
Only once before has a form of this group been reported from the Guahyba- drainage, viz. D. martensi Von Ihering, from Taquary and Santa Cruz (probably from tributaries of the lower Jacuhy (See Von Ihering, 1893, p. 102). But these are not typical martensi, the original of the latter being “ probably from Sao Paulo).””” The dimensions of these specimens from Rio Grande do Sul given by Von Ihering agree fairly well with those of the real U. martensi, chiefly the relation of height to length: 49 pr. ct. in two specimens from Rio Grande do Sul, 49 pr. et. (text) or 53 pr. ct. (figure) in martensit. But these dimensions do not agree with my speci- mens, where the height ranges from 55 to 64 pr. ct. and never falls as low as in martenst. The diameter of martensi is also greater than in my material. For this reason I am compelled to describe my shells as a new species.
Description of Shell—Of medium size (maximum length 80 mm.), moderately solid, rather thin when young, compressed (diameter 26 to 33 pr. et. of length), subelliptical, subovate, or subtrapezoidal, moderately elongated (height 55 to
Tt is much to be regretted that the type-locality is not better known. It is very doubtful, on
account of the great similarity of these forms, whether the real martensi can ever be positively identified.
492 MEMOIRS OF THE CARNEGIE MUSEUM.
64 pr. ct. of length). Valves not gaping. Dorsal margin nearly straight or very gently convex, passing gradually into the anterior margin (rarely forming an in- distinct angle). Posterior upper margin forming a blunt angle with the posterior margin or passing gradually into it. Posterior margin obliquely descending, more or less curved, and curving more strongly into the lower margin, without forming a distinet posterior angle. The lower posterior end of the shell is rounded, but hardly biangular. Lower margin in young specimens gently convex, in older ones it is rather straight in the middle. Anterior end of shell not, or very little, narrower than the posterior end. The shell is thus rather straight and not oblique.
Valves only slightly convex, rather flattened upon the sides, with the posterior ridge very indistinct and broadly rounded. Posterior slope somewhat compressed. Greatest diameter near the middle of the shell, but not posterior to it. Beaks not swollen, and not elevated, located at from 20 to 28 pr. ct. of the length. Beak- sculpture consisting of about fourteen to sixteen rather sharp and fine, short, radial bars, 5 to 7 mm. long, those upon the posterior ridge hardly longer than the rest, and none of them uniting in the middle of the shell with their lower ends. No distinct granulations present, but sometimes there are a few irregular oblique wrinkles upon the posterior slope near the beaks. A short, narrow lunula in older shells.
Epidermis in young specimens rather smooth and shining, but with fine, ir- regular, concentric striae, nowhere lamellar. In old specimens it is less smooth, chiefly on the posterior slope and toward the lower margin with more crowded and rougher striz. Crinkled radial lines hardly indicated upon the shell. Color in young specimens greenish bronze or brownish, sometimes with indistinet brown- ish concentric bands, in older shells greenish tints disappear, and the epidermis is dull brown or blackish brown. ;
Hinge-line gently curved. Ligamental sinus over the middle of the lateral tooth or slightly behind it, in older shells over its last third. Laterals curved, one in right, two in left valve, somewhat rough posteriorly. Pseudocardinals normally two in each valve. In young specimens those of the right valve are obliquely and forwardly descending, compressed, the anterior low and narrow, the posterior higher and a little thicker, crenulated. In older specimens the posterior is thicker and becomes generally more triangular and stumpy. Those of the left valve of _young specimens are also compressed, but subtriangular and not very long, crenu- lated, the anterior one larger than the posterior. In old shells these two teeth are more stumpy, triangular, and not compressed. In rare cases reductions take place, chiefly with regard to the anterior tooth of the right valve, or with regard to the
ORTMANN: SOUTH AMERICAN NAIADES. 493
posterior in the left valve, which may become very small, and sometimes an ac- cessory third posterior tooth may develop in the right valve. Thus the pseudo- cardinals are quite variable in number, shape, size, compression, and development of rugosities.
Cavities of shell and beaks shallow. Nacre blueish white or white, often discolored and with lurid tints (grayish purple) toward the beak-cavities, iridescent posteriorly.
Anterior adductor-sear distinct and impressed, subtriangular. Anterior retractor-scar separated from it, small, round, deeply impressed. Anterior pro- tractor-scar connected with adductor-scar, forming a posterior process of it. Pos- terior adductor-sear shallow, subtriangular, with an upper process formed by the posterior retractor-scar. Pallial line distinct. Dorsal muscle scars a few, lying in a line in the beak-cavity.
MEASUREMENTS.
No. Sex.| Length. Height Diameter. Beaks. Figure
11...| o@ 40.5 mm. |24 mm. =59 pr. ct. of L..12 mm. =30 pr. ct.of L..at 11 mm. =27 pr. ct. of L.
16...| co |43 Pr 27 = =6a ee oe 11 =26 re
Serle ie seo | 130) 04 sf oe 11 oe a WOPISEXCRGNGV tips.
29... .| co 162 “134 oD. 14 fo) st Pl. XXXIV, fig. 5. Ar el a Fees pe en ae (PI. XXXIV, fig. 7. eoea3| 2 (3 pon : = | Pl. XXXYV, fig. 1.
Anatomy.—Anal opening slit-like, shorter than branchial opening, closed above; closed part about twice as long as open part, no supra-anal formed. Bran- chial opening separated from anal by a solid connection of the inner mantle-edges. Inner edge of anal almost smooth, that of branchial with distinct papille. In front of the branchial the mantle-edges are unconnected. Palpi subtriangular, a little longer than wide, their posterior margins connected at base.
Gills long, the outer subtriangular, wider in the middle than the inner. The inner not triangular, wider than the outer anteriorly, its anterior end attached to the space between anterior end of outer gill and palpi, and in contact with the posterior base of the latter. Inner lamina of inner gill entirely connected with abdominal sae.
Non-marsupial gills (Pl. XLV, fig. 1a) with few, scattered, irregularly dis- posed interlaminar connections. Marsupium (Pl. XLV, fig. 1b) located in the inner gills, occupying only a part of them, at and in front of the middle; at anterior end about one-fifth of the gill remains non-marsupial, and at the posterior end about two-fifths, so that the marsupium occupies two-fifths of the gill, the second and third from the anterior end. This location of the marsupium is constant for
494 MEMOIRS OF THE CARNEGIE MUSEUM.
the species (observed in seventeen females), with the only qualification that in young females the marsupium is smaller (less than two-fifths of the gill) but has a similar position.
Marsupial part formed by interrupted septa (Pl. XLV, fig. 1b; Pl. XLVII, fig. 6). The arrangement into septa is distinct, but they are frequently inter- rupted, and their solid portions are short. In the middle of the marsupium they are at the utmost six to eight times as long as thick, and in its other parts they are very short. A distinet quincuncial (reticulate) arrangement is not seen. Toward the margin of the gill the septiform arrangement is again more distinct.
The embryos fill the marsupium in an irregular mass, and the charged mar- supium, is moderately swollen. Placents are not formed.
Glochidium (text-fig. 4b, p. 469, observed in six specimens) subtriangular, slightly oblique, anterior and posterior margins convex, converging to a point, anterior margin longer than the posterior. Hooks present, of the Hyriine type, with the S-shaped curve. L. 0.27 to 0.28 mm.; H. 0.27 to 0.28 mm.; L. of hook: 0.09 mm.
It should be remarked that in one female the glochidia were immature, and no hooks could be seen; three females had only eggs, and one female was in the act of discharging. The dates of collecting, Jan. 26 and 29, should be recorded as showing the breeding season (midsummer of Southern Hemisphere).
Remarks on the Specific Characters.—As will be seen from the descriptions of the following species, D. simillimus, D. vicarius, and D. decipiens are very closely allied to the present species in the shape of the shell; vicarius can be readily dis- tinguished by the biangular posterior ridge, and D. simillimus is a smaller shell. In other respects it is almost impossible to distinguish these species by the shell alone. The few obscure, differentiating characters will be pointed out below under the respective species. However, the investigation of the soft parts has shown that there are interesting differences chiefly in the location of the marsupium, In the present species, the marsupium is most fully developed, occupying a rather large part of the inner gills, slightly gravitating toward the anterior end. In the following species (chiefly simillimus and vicarius) it will be seen that this tendency is increased, and the marsupium becomes smaller, and is being shifted more dis- tinctly forwards (Compare pl. XLV, fig. 1b, with Pl. XLV, figs. 2b, and 3). It will also be seen that there are slight differences in the glochidia with regard to obliquity and size.
ORTMANN: SOUTH AMERICAN NAIADES. 495
9. DIPLODON SIMILLIMUS Ortmann, sp. nov.
Shells: Pl. XX XV, figs. 38, 4, 5,6. Anatomy of gills: Pl. XLV, fig. 2. Glochidium: Text-fig. 4c, p. 469.
Type-locality.—Rio Nhundiaquara, Morretes, Parans, Brazil (J. D. Haseman coll., January 3, 1909). Type-set: Carn. Mus. Cat., No. 61.9250. Thirty-two specimens with soft parts, males, barren and gravid females with glochidia.
About a dozen additional specimens belonging to the same original lot have been studied. The locality is in a small coastal stream emptying into the Bay of Paranagua.
No shells have ever been reported from the region of Paranagua Bay in Paranda. However, from a little less than one hundred miles to the south, in Santa Catharina, near Barra Itapocu (I believe that the Rio Itapoca, as given by Marshall, stands for this), Diplodon santamarie Simpson has been described (Simpson, 1914, p. 1270; and Marshall, 1917, p. 386, Pl. 52, fig. 6; Pl. 55, figs 1-4). This is founded upon three specimens only, and resembles our species to a degree. But judging from description and figures, it is somewhat larger (max. 63 mm.), longer (H. 52-59 pr. ct. of L.), and the hinge has the posterior tooth of the left valve missing. Since nothing is known of the anatomy of D. santamarie, and the locality is not the same, it would be rash to unite our specimens with this species.
Our species also is much like U. martensi Von Thering, and might fall under this according to Von Ihering’s conception. But it cannot be united with it on account of the different dimensions. While in martensi the height is said to be from 49 to 57 pr. ct., our specimens are mostly less elongated, with the height from 53 to 66 pr. ct. of the length. In D. martensi the diameter is 32 to 35 pr. et., in the present species from 26 to 37 pr. ct. (this agreeing better with martensi than with imitator). But in the absence of exact localities for martensi and any knowledge of its anatomy, and in view of the general resemblance of all of these shells, it is impossible to identify our shell with any previously described, and no other alterna- tive exists, except to describe it as new.
In the characters of the shell D. simillimus is very close to D. imitator. The description of the latter species would fit it very well, and I shall here only emphasize the distinguishing characters.
1. D. simillimus is a smaller shell (max. length 61 mm., as against 80 mm. in imitator).
2. The trapezoidal shape, with an angle between the upper and the posterior margins, is seen here only in very young shells. In older shells these two margins form a rather regular curve.
496 MEMOIRS OF THE CARNEGIE MUSEUM.
3. The lower margin of the shell of D. simillimus is frequently more nearly straight, so that the shell of older specimens appears more humped.
4. Color of epidermis with hardly any green tints, but light to dark bronze- brown, in old shells brown-black.
5. Pseudoeardinal teeth never stumpy, chiefly in left valve, but always com- pressed, although the two of the left valve and the posterior in the right are rather thick in old specimens.
MEASUREMENTS.
No. |Sex Length. | Height. Diameter. | . Beaks. Figured.
5... 36.5 mm. 21.5 mm. =59 pr. et. of L.| 9.5 mm. =26 pr. ct. of L. at 8.5 mm. =23 pr. et. of L. DOS ANS KODE ae oe “ =61 < 11 =29 te 10 ei on
18...) 9 | 42.5 22.5 =53) a 11 =26 ~ 95 “ =22
11...) oc! | 45 27 =60 13 = 9 “ =20 Pl. XXXYV, fig. 3. 233. 21 O59 39 =66 22 =f 14 =24 os 22 |ole1 “ |37 =61 20.5 | [Sulla at pean ar peaee Pl. XXXV, fig. 5.
The characters of the shells of this species are rather poorly marked, and can be ascertained only by the examination of extensive material. I have discovered, however, that there are important and constant differences in the anatomy.
Anatomy.—Fully agreeing with that of D. imitator, but the marsupium is different (See Pl. XLV, fig. 2). It is located in the anterior part of the inner gill, entirely anterior to its middle, and extending forward to within a short distance of the anterior end of the gill, so that anteriorly less than one-tenth of the gill is non-marsupial, while posteriorly fully the posterior half of it is non-marsupial. The interlaminar connections form interrupted septa, but the septiform structure is less distinet than in D. imitator, and more of a reticulated (or irregular quin- cuncial) arrangement is evident. This structure of the marsupium is the same in all females investigated (altogether twenty individuals), and only in young ones is the marsupial part smaller.
The Glochidium (Text-fig. 4c, p. 469) differs from that of D. imitator in being more distinctly oblique, and being longer than high. L. 0.28 mm., H. 0.24 mm. There are hooks of the same type, which are about 0.10 mm. long.
My specimens were collected on January 3, which date indicates the breeding season, probably its middle, for eight females had only eggs, while nine had glochidia, in part not mature, and with the hooks yet unformed.
The remarkable fact brought out by the study of the anatomy is that, while the species is extremely hard to distinguish from imitator by the shell, it has at least two anatomical characters (marsupium and glochidium), which are very well marked
and constant.
ORTMANN: SOUTH AMERICAN NAIADES. 497
10. DrpLopon vicartus Ortmann, sp. nov.
Shells: Pl. XXXYV, figs. 7, 8; Pl. XXXVI, figs. 1, 2. Anatomy of gills: Pl. XLV,
fig. 3. Glochidium: Text-fig. 4d, p. 469.
Type-locality—In creeks, Aqua Quente (eight miles from Iporanga), Sao Paulo, Brazil, tributaries of Rio Ribeira (J. D. Haseman coll., November 27, 1908). Type-set: Carn. Mus., Cat. No. 61.9251; fifteen specimens all with soft parts, males, barren and gravid females.
Additional Locality.—Rio Ribeira, Iporanga, Sao Paulo, Brazil (J. D. Haseman coll., December 1. 1908). One specimen, male, with soft parts.
Only one species of Diplodon has been described from the Rio Ribeira; this is D. mimus Simpson (1914, p. 1249; Marshall, 1917, p. 383, PI. 51, fig. 3) from Iguapé, Sao Paulo, at the mouth of the river. As has been pointed out above (p. 486), this species might possibly be D. granosus. But on the other hand a few particulars agree with the present species, as, for instance, the biangulation of the posterior end and the general resemblance to Elliptio complanatus mentioned by Simpson. Yet our shells cannot be this, because they are larger, less convex, and have dif- ferent dimensions. D. mimus is smaller (max. 45 mm.), and has, according to the measurements given, a considerably higher (60 to 76 pr. et.) and more swollen (D. 33 to 46 pr. ct.) shell, while D. vicarius has the height only from 52 to 63 pr. et. and the diameter from 26 to 31 pr. et., and is a good deal larger (L. 53 to 68 mm.). Of course, our form may fall under martensi Von Thering, but for the same reasons as in the case of the two preceding species, it cannot be called by this name, and thus we must describe it as new.
It may perhaps be that D. vicarius, of which we do not know the beak-sculp- ture, is the older stage of the young specimens recorded from Iporanga as D. granosus (See p. 485). The shape and dimensions agree fairly well, but the size is very different, the maximum length of granosus being only 29 mm. and no inter- mediate specimens between these and minimum length of vicarius (53 mm.) are at hand. Thus the question must remain unsettled. D. granosus from other localities is also always much smaller than vicarius.
This species is also extremely similar to D. imitator and D. simillimus. Yn size it stands between them (max. length 68 mm.). The outline is also subtrapezoidal or subelliptical, and the shell is quite compressed, resembling the shape of Elliptio complanatus of the United States. The lower margin in the shells before me is always rather straight, but I have not very young specimens. However, from the growth-lines it is seen that young shells must have had a gently curved lower margin. In none of my specimens is the beak-sculpture preserved.
498 MEMOIRS OF THE CARNEGIE MUSEUM.
Characters of the Shell—The description of D. imitator might also serve for this species. However, the following differences are noticeable:
1. Posterior ridge of shell broad, and more or less distinctly biangulate, pro- ducing a biangulation also of the posterior end of the shell. This is the most striking character of the shell. In both D. imitator and simillimus the posterior end of the shell is evenly rounded without any trace of angulations.
2. Epidermis not so shining as in the two other species, which is due to the development of additional fine concentrie wrinkles, which are irregular and best developed upon the posterior slope and near the lower margin. Upon the disk they form indistinct radiating lines (or narrow bands), which are obsolete in the two other species. Color of epidermis of vicarius lighter or darker brown, without green tints, and without bronzy lustre.
The hinge-teeth are generally of the type of D. simillimus, 7.e., they do not become stumpy in old shells. They are, however, very variable, and much cut up, more so than in the two other species, and the posterior pseudocardinal of the left valve frequently is quite small or rudimentary. The nacre is whitish, but often inclines to lurid tints (purplish gray).
MEASUREMENTS.
Height. Diameter. Beaks. Figured.
No. ISex.| Length. |
: 13...| | 53. mm.31_ mm.=58 pr. ct. of L./15.5 mm. =29 pr. ct. of L..at 12 mm. =23 pr. et. of L.| Pl. XXXV, fig. 7. 9...| 9 | 63.5 “ 33.5 “ =63 os 14 =26 iW 12 “ =22 | Pl. XXXVI, fig. 2. 10:2 3/9! 57 “34.5 “ =61 i: 16 28 MS) os, =20 - | 1...| 9 | 58.5 33.5 =5i/ % 15 i 15 Sate
o'| 67 35 =52 9 =23 ) “ =2 yyy
| | Pl. XOEXVI, fig. 1.
No sexual differences in the shell.
Anatomy.—I have eight females, six of which are gravid; three with eggs, three with glochidia (immature in one). The structure of the soft parts is exactly as in the preceding species, except that of the marsupium (Plate XLV, fig. 3). As in D. simillimus, the marsupium is located in the anterior portion of the inner gill, anterior to the middle, but it occupies a still smaller part, and does not extend so near to the anterior end, and does not reach the middle of the gill. Anteriorly about one-ninth of the gill is non-marsupial, and posteriorly over half of it. The interlaminar connections are arranged in interrupted septa, the septiform arrange- ment being most evident anteriorly and in the middle of the marsupium, while posteriorly the arrangement is reticulate (indistinetly quincuncial).
The glochidium (‘Text-fig. 4d, p. 469) is similar to that of D. simillimus, oblique, but slightly longer. L. 0.30, H.0.24mm. There are hooks of the same type, which
ORTMANN: SOUTH AMERICAN NAIADES. 499
are at least 0.09 mm. long. In one specimen with immature glochidia no hooks could be seen.
In this case also the breeding season falls in the winter months of the northern hemisphere, but apparently a little earlier than in the other species (end of No- vember).
Thus this species has anatomical characters of its own, chiefly observable in the size and location of the marsupium. It agrees most closely with D. simillimus.
3. SPECIES FROM THE DRAINAGE OF THE UPPER PARANA.
I have material belonging to the chilensis-group from the following tributaries of the Parana: Rio Iguassti, Rio Tieté, and Rio Grande in Sao Paulo. A form from the Iguassti is noticeably very closely related to the three species from the coastal streams just described. We should bear in mind that the head-waters of the Iguassu, from which this form comes, are in close proximity to those of the Rio Nhundiaquara and Rio Ribeira on the eastern watershed.
11. DipLopoN DECIPIENS Ortmann, sp. nov.
Shells: Pl. XXXVI, figs. 3, 4, 5, 6. Anatomy of gills: Pl. XLV, fig. 4. Section of gills: Pl. XLVI, fig. 7. Glochidium: Text-fig. 4e, p. 469.
Type-locality— Creek, tributary to the Rio Iguasst, Serrinha, Paranda, Brazil (J. D. Haseman coll., December 23. 1908). Type-set: Carn. Mus. Cat. No. 61.9253, thirteen specimens, males, barren and gravid females, all with soft parts.
No Naiades have hitherto been known from the river-system of the Iguassu, and, as in the preceding species, none of the names of species which may occur here, ean be applied to our specimens with any degree of certainty. Therefore we in- troduce this form as a new species.
In the shape of the shell this species is very close to the three preceding, es- pecially imitator and simillimus, which it resembles in its subelliptical or sub- trapezoidal outline. The latter shape is seen chiefly in younger specimens, while older ones become more subelliptical. The posterior ridge and the posterior end are never biangulate as in vicarius. In the glossy epidermis, D. decipiens also agrees better with imitator and simillimus, and the radial lines formed of fine wrinkles are poorly developed. The general description given for imitator might be repeated for this species, and the beak-sculpture in particular has the same character. Nevertheless the following peculiarities should be mentioned as diagnostic:
500 MEMOIRS OF THE CARNEGIE MUSEUM.
1. The beak-sculpture has a smaller number (ten to twelve) of radial bars, the bars having the same length (5 to 7 mm.)
2. The outline of the shell is more frequently subelliptical, often rather regularly so, with both upper and lower margins almost equally curved. In some specimens, indeed, the lower margin is more nearly straight, but this never causes a distinctly humped shape (so often seen in simillimus). No trace of biangulation posteriorly.
3. Epidermis in young shells bronzy-brown or bronzy-green; in older ones it becoming a deep chestnut-color, inclining partly to blackish.
In size this species stands between imitator and vicarius (maximum length 73 mm.). The relative dimensions are very much like those of the other three species. The hinge-teeth most resemble those of simillimus, never being stumpy, as in imitator; they are not much cut up; and the posterior pseudocardinal in the left yalve has a strong tendency to disappear, being sometimes entirely missing. The nacre is whitish, but generally lurid (purplish gray) in the cavity.
MEASUREMENTS.
Height.
Length.
No. |Sex. Diameter. | Beaks. | Figured. a...|o' | 26mm. |14.5 mm. =56 pr. ct. of L.. 6 mm. =23 pr. ct. of L.'at- 6.5 mm. =25 pr. ct. of L.|
c OF poDu es 20:5 ** 59 Wane Ew ee) oy | Ulelsy I SPAN 2 Pl. XXXVI, fig. 6. 3 2 | DS Boo) =56) y 18250, “= 32 - | 15 26) y Pl. XXXVI, fig. 5. 4 op eee 4 36.5 =54 200i see — ol sg | 15 “ =22 oe Pl. XXXV, fig. 3. LOS | 70 38 ole 25 wi Sale pe 16 =23 Me
aE ayer 40 =55 25 Set! : 16 ‘ -=22 3 Pl. XXXVI, fig. 4.
No sexual differences in the shell.
Anatomy.—Aside from three very young specimens, which probably are a male and two females, I have six males of good size, and one barren female, a gravid female with eggs, and two females with glochidia (in one of them immature).
The anatomy is similar to that of D. imitator, simillimus, and vicarius, but the marsupium (Pl. XLV, fig. 4b) is quite different, essentially differing from that of the other species. It is located in the middle of the inner gill, leaving between one- fifth and one-fourth of the gill at the anterior end, and about one-third of it at the posterior end non-marsupial. Thus the marsupial portion is rather large, and more of it is anterior rather than posterior to the middle. This location comes nearer to what is seen in D. imitator, than in the other two species. The most striking character, however, which we have not before encountered, is that the interlaminar connections of the marsupial part are not interrupted, but form continwous septa, running from near the base of the gill close to the margin. These septa are heavy and stand very closely, forming regular, isolated water-tubes (See section on PI. XLVII, fig. 7). The whole of the marsupium has this structure, and no inter-
ORTMANN: SOUTH AMERICAN NAIADES. 501
rupted septa are seen anywhere. That these septa are only a modification of the scattered interlaminar connections is shown by the fact that the latter are found in the non-marsupial gills (See outer gill on Pl. XLV, fig. 4b, and gills of male, Pl. XLV, fig. 4a). All four of my large females show this arrangement, and in the two young ones traces of the beginning of this structure are seen. The em- bryos fill the water-tubes in loose masses, not forming distinct placentz.
Glochidium (Text-fig. 4e, p. 469) much like that of D. vicarius, oblique, L. 0.31, D. 24 mm., with hooks, 0.09 to 0.11 mm. long. One specimen has immature glochidia, but rudimentary hooks can be seen. The date of collecting (December 23) gives a hint as to the breeding season.
While this species is very close to the three preceding in the characters of the shell, it has differences in the soft parts, which are very striking. The structure of the marsupium is extremely interesting, and there is no doubt that it must be regarded at least as of specific value, representing a high specialisation of this organ.
12. DreLopon pautista (Von Ihering) (1893).
Anatomy of gills: Pl. XLVI, fig. 1. Section of gills: Pl. XLVIII, fig. 1; Glochidium: Text-fig. 4 f. p. 469.
Unio paulista Von Txertna, 1893, p. 93, Pl. 4, fig. 7. Diplodon paulista Stupson, 1900, p. 873; 1914, p. 1229.
Type-locality—Piracicaba, Sao Paulo, Brazil; according to Nehring (1893, p. 166) in Rio Piracicaba Mirim.
New Localities.—Rio Tieté, Mogy das Cruzes, Sio Paulo, Brazil (Headwaters of R. Tieté) (J. D. Haseman coll., July 19, 1908) males, barren and gravid females, originally twenty-five in the lot, all with soft parts. Rio Tieté, Sapina, Sao Paulo
bo
(Exact location unknown, but must be near city of Sado Paulo) (J. D. Haseman coll., July 23, 1908) one male with soft parts. Creek, tributary to Rio Mogy Guasst, Mogy Mirim, Sao Paulo, Brazil (tributary to Rio Grande and Paranda) (J. D. Haseman coll., August 7, 1908) males and gravid females, originally seven- teen specimens in the lot, all with soft parts.
A detailed description of five specimens has been given by Von Ihering, but the specific characters have remained obscure. The specimens before me agree very well with this description, but it should be noted that the two forms of the shell supposed by Von Ihering to belong to the male and female, do not represent sexual differentiation, but simply individual variations. The more regularly ovate outline, believed to belong to the male, is in fact rather rarely well-developed,
502 MEMOIRS OF THE CARNEGIE MUSEUM.
while the other, somewhat more oblique, is more frequent; but both pass insensibly into each other. According to my material younger specimens are more apt to exhibit the more regularly ovate outline.
This species is not very closely allied to those mentioned on the preceding pages, but represents a somewhat different type, and inclines towards the ellipticus- group by its often slightly oblique shell. It is much smaller than imitator, vicarius, and decipiens, and also does not attain the size of simillimus. The largest specimen before Von [herimg was 57 mm. long, while my largest falls even short of this (45 mm. from Rio Tieté, 51 mm. from Mogy Mirim). The subtrapezoidal shape is only distmet in very young individuals; in larger specimens it becomes subelliptical or subovate, generally a little higher posteriorly and slightly oblique. The ground- color of the epidermis is greenish-olive, and never distinctly brownish, although old specimens may become blackish green. The beak-sculpture consists of fine, sharp, and short radial bars, the number of which is sixteen to eighteen, the median pair having a tendency to unite at the lower ends.
The characters of the inside of the shell have been well described by Von Ihering. The left valve has generally only one pseudocardinal, but sometimes there is a trace of a second posterior one. It should be noted that, as described by Von Ihering, the posterior retractor-scar is separated from, and stands above the posterior adductor-sear, but not always, as in some eases it is connected with it, and this may be different, even in the right and left valve of the same individual.
MEASUREMENTS.
No. Sex. Length. | Height. Diameter. Beaks. Von Ihering... .. 48 to 57mm. 59 to 65 pr. et. of L. 31 to 35 pr. et. of L. 22 to 26 pr. et. of L. Mogy das Cruzes ? 15.5 mm. 8.5 mm. =55 * 4 mm. =26 = at 4 mm.=26 x
LS Aravs feet te) 27 ce 16 53!) 8 0) & 6 ‘ =22 oes LOK Se ow | 38 Ne 22 Fs 4 Gp ee sly) cS RD e s (=P 10) as 2 weer svar 2 38 SS 24 Os) 4 14 Sor o 10 Si 6 a
(SAG CEI fof 41 Ce 21GOM ee Ot 14 =34 i 10 “ =24 LAY era ss fot 45 a 27 “ =60 ss 16 " =36 - 11 "== 24 sy Mogy Mirim
ees ot |) 335) $65 WD GS) A } 14.5 ‘ =41 : 10 « =29 s ee ote Oo 49 | 23.5 =60 13 co =a's} xf eae De apel = ee oc fr o 40:5: | 24 DS 15 Says a ' LD —
YR fou 2 : 28 OF : (eS, | As 11.5 “ =27 ‘< Been 2 48 es 29 G0) u 16:55“ =34 13 =a ve Seeks | of 51 |, 82 “ =63 ‘ [1G S35 s 13 “ =26 os
My two sets from the Rio Tieté and from a creek tributary to Rio Mogy Guassu, differ slightly from each other. In the former the size is a little smaller, and the young shells are rather more elongated, thus rendering the average height less. The diameter is also not so great in the shells from the Tieté. The measure-
ORTMANN: SOUTH AMERICAN NAIADES. 503
ments of the dimensions of the two sets largely overlap, and they agree very well with those given by Von Ihering. Of course, my material being more plentiful, the range of the dimensions is wider.
An apparently allied form is D. suppositus Simpson (1914, p. 1245; Marshall, 1917, p. 385, Pl. 51, fig. 2; Pl. 54, fig..1-4).1° The type is, according to Marshall, from “Paranda, Brazil,’ and other specimens are from Rio Tieté, Sao Paulo, and other localities in southern Brazil. This is also a comparatively small form, but, according to the measurements given, it is more elongated, the height being in the type 52 pr. ct. of the length (53 pr. et. according to Marshall), and in another specimen 58 pr. ct. This differs somewhat from our specimens (58 to 67 pr. ct.) and from Von Ihering’s paulista (59 to 65 pr. et). In addition suppositus has a chestnut-bronzy epidermis, while it is greenish in paulista. Thus the two forms do not agree.
Anatomy.—I have investigated the soft parts of all of my shells, and there were altogether, aside from several small ones, where the sex could not be ascer- _ tained, nineteen males, six barren females, five gravid females with eggs, and seven gravid females with glochidia.
The anatomy in general is like that of the genus Diplodon. But in this case again the marsupium (Pl. XLVI, fig. 1) in its size and location shows specific peculi- arities, exhibited by all of my females. It agrees with the preceding species in the fact that it occupies only a part of the inner gill, and in the slightly anterior location. But the marsupial part is very small, occupying about one-fifth or one- fourth of the length of the gill, leaving a considerable portion non-marsupial at the anterior end, and half or nearly half at the posterior end. The figure on PI. XLVI, fig. 1, represents a specimen with a rather small marsupium; generally it is a little larger. Thus the marsupium is in the middle of the gill, and slightly in advance of the middle. When charged, it forms here a rounded or oval swelling, rather distant from the base, and extending not quite to the edge. There is a slight variation in the specimens from the two main localities. In those from Mogy das Cruzes it is distinctly in front of the middle; in those from Mogy Mirim more median, but this difference is very slight and not always distinct. The inter- laminar connections form interrupted septa (also seen in the section, Pl. XLVIII, fig. 1), and there is no distinctly reticulate arrangement.
Glochidium (Text-fig. 4f, p. 469) rather large, subtriangular, oblique, with hooks. L. 0.32, H. 0.27 mm.; hooks about 0.10 mm. long. (Thus the glochidium is slightly larger than in any of the preceding species.)
16 The name suppositus was first given by Von Ihering (1893, p. 102) without description, from Rio Grande (Upper Parand, boundary between Sao Paulo and Minas Geraes).
504 MEMOIRS OF THE CARNEGIE MUSEUM.
4. Group oF Diplodon charruanus.
Shell not compressed, but rather evenly convex and often considerably swollen when old. Outline subelliptical or somewhat subtrapezoidal, elongate, more or less poimted behind, straight, and not oblique. Beak-sculpture simple, with narrow, straight, uninterrupted bars, restricted to the region of the beaks. A few of these bars may be joined at their lower ends.
The chief character by which this group is distinguished from that of chilensis is the absence of a flattening of the valves upon the sides. The valves are generally evenly convex, and in consequence of this, at least in larger specimens, the shell appears as more swollen. In other respects, the shelbis similar in outline and other characters, except that the posterior end is often produced into a distinet blunt point, which may be more or less elevated above the base-line. There is no dis- tinct tendency of the posterior portion of the shell to be higher than the anterior, and thus the shell is not oblique. A posterior ridge may be distinct or indistinct.
The first species, D. parallelopipedon, (Lea) froms a transition toward the group of chilensis.
The metropolis of these forms is in the system of the Rio de la Plata, but they are also found in the coastal streams of southern Brazil (Rio Grande do Sul).
13. DIPLODON PARALLELOPIPEDON (Lea) (1834).
Unio parallelipipedon D’Orsiany, 1848, p. 609; Corst, 1900, p. 447, fig. 30. Diplodon parallelipipedon Simpson, 1914, p. 1275.
Unio parallelopipedon Von Martens, 1894, p. 164; Piuspry & Rusu, 1896, p. 30. Diplodon acutirostris (LEA) Simpson, 1914, p. 1276.
Diplodon parallelipipedon and acutirostris HAaAs, 1913, pp. 22, 28, 52, 53.
Type-locality—Rio Parana, Province of Corrientes, Argentina.
Other Localities—Arroyo del Rosario, Uruguay (to La Plata) (D’Orbigny) ; Arroyo de las Vaceas, Uruguay (Corsi); Rio de la Plata, Colonia, Uruguay (Pilsbry & Rush); Rio Uruguay, Salto Oriental, Uruguay (Haas); Swamps of Rio Parana from Buenos Aires to above Corrientes, Argentina (D’Orbigny); Rio Paraguay (Von Ihering, 1893, p. 119); Paraguay (Von Martens); Rio de San Miguel, Prov. of Chiquitos, Bolivia, (D’Orbigny).
The last locality deserves special attention, since it is in the Amazons-drainage; but it is in the most southern extremity of it, close to the divide toward the Para- guay. It certainly requires confirmation.
New Localities—Rio Uruguay (in mud), Uruguayana, Rio Grande do Sul,
ORTMANN: SOUTH AMERICAN NAIADES. 505
Brazil (J. D. Haseman coll., February 5. 1909). One specimen, male, with soft parts. Pond along Rio Negro, Santa Isabel, Uruguay (J. D. Haseman coll., February 11. 1909). Four complete shells, and five isolated valves; of two of these soft parts, males.
Distribution.—-Drainage of the Rio de la Plata from its mouth and its tributaries in Uruguay and southern Brazil up to the Rio Paraguay in Paraguay. Also reported as crossing the divide, and going into the headwaters of the Amazons in Bolivia. Not known from the Paranda above Corrientes.
This is a species easily recognized by the elongated-subtrapezoidal shape, with the upper and lower margins nearly parallel, by the anterior position of the beaks, by the rather swollen shell and distinet (although rounded) posterior ridge. The sides of the disk are rather flattened, and in this respect this species resembles the chilensis-group, intergrading with it to a degree. My specimens are somewhat variable in shape, being longer or shorter. None of them shows the beak-sculpture, since the beaks are badly eroded in all, except in the specimen from Uruguayana, where they are only a little eroded, and consequently a little more elevated. But here also no beak-seulpture can be seen. It must occupy only a very short space near the beaks, hardly more than 5mm. The specimen from Uruguayana has the nacre suffused with red (already mentioned by D’Orbigny).
There is not the slightest question that U. acutirostris Lea is an old, much eroded, and somewhat distorted specimen of this species. Haas has already sug- gested this.
Anatomy.—I have only the soft parts of male specimens, but Lea has already described them in the ease of his D. acutirostris, and has furnished at least some information about the marsupium.
Judging from my material, the anal opening is closed above without forming a supra-anal. Closed part over five times as long as the open, the latter slit-like, short, shorter than the branchial opening. Inner edge of anal indistinctly crenu- lated. Anal and branchial separated by a solid bridge, running a certain distance inward. Branchial with fine papillae, about three times as long as anal. Mantle- edges not united in front of it. Palpi subtriangular, lower margins convex, pos- terior margins connected for about one-fourth or one-third of their length.
Gills long and rather narrow. In their posterior part they are of equal width, or the outer one is slightly wider; anteriorly the inner is much wider. ‘The outer is considerably narrowed anteriorly, its anterior end being situated near the highest point of the line of the attachment of the mantle. The inner gill has a straight margin in the middle, and anteriorly it is only slightly narrower, its anterior end
506 MEMOIRS OF THE CARNEGIE MUSEUM.
being immediately behind the palpi. Inner lamina of inner gills entirely con- nected with abdominal sac. Structure of the gills in the male as usual, but the interlaminar tissue is unusually well-developed, forming a rather thick layer chiefly on the inside of the primary limb of the gill; and it has, as usual, short, interrupted interlaminar connections, elongated in the direction of the gill-filaments, which in the middle of the gill are few and far apart, while they are a little more frequent near the ends.
Lea describes the marsupium (of acutirostris) as occupying nearly the whole length of the inner gill, but no information is given as regards the finer structure.
Color of foot brown or blackish in the distal part, otherwise the soft parts are whitish.
14. DreLopon cHARRUANUS (D’Orbigny) (1835). Glochidium: 'Text-fig. 4g, p. 469.
Unio charruana D’OrsrGny, 1843, p. 606, Pl. 71, figs. 8-11; Pinssry & Rusu, 1896, p. 81; Corst, 1900, p. 447, fig. 31.
Unio faba (as form of charruana) D’OrBiaNy, 1843, p. 606 (text), as rhwacoica, Pl. 71, figs. 12-14 (in tabula per errorem, see Explanation of plates, p. 704). Unio rhuacoica D’OrsBiIGNyY, 1843, p. 606, Pl. 69, figs. 4, 5; Corst, 1901, p. 450,
fig. 33. Unio aethiops Lea, Obs., X, 1868, Pl. 41, fig. 285 (jwv.). Unio parcus Lea, Obs., XII, 1869, Pl. 33, fig. 77 (juv.). Diplodon rhuacoicus, charruanus, ethiops, parcus, Sumeson, 1914, pp. 1242, 1243,
1247, 1256. Diplodon hidalgot Haas, 1916, pp. 18, 49, Pl. 1, fig. 1. Diplodon parcus Haas, 1916, pp. 16, 49. Diplodon fortis MARSHALL, 1917, p. 382, Pl. 52, figs. 1-4.
Type-locality—Small streams from Maldonado and Montevideo to Las Vacas, Uruguay (‘‘ Banda Oriental’’).
Other Localities—Lake Potrero, Maldonado, Uruguay (Pilsbry & Rush); Rio Canelon Grande, Montevideo (D’Orbigny, rhuacoica); Dep. Canelones, Uru- guay (Corsi); Rio Miguelete, Uruguay (Haas, hidalgo’); Rio Negro, Tacuarembo, Uruguay; correctly $8. Fructuosa, on Rio Tacuarembo, tributary to Rio Negro (Marshall, fortis); Uruguay River (Lea, ethiops).
New Locality.—Pond along Rio Negro, Santa Isabel, Uruguay (J. D. Haseman coll., February 11. 1909). About twenty-five specimens, seventeen with soft parts, including males and gravid females.
ORTMANN: SOUTH AMERICAN NAIADES. 507
Distribution.—Small streams of the “Banda Oriental” in Uruguay, from Maldonado westward, and also in the Rio Negro and the Rio Uruguay.!7
An extremely variable form, of which I possess a good set from one locality, undoubtedly representing young and adult specimens of the same species, so that I am able to give a rather full account of it. It is very evident that different in- dividual phases have been previously deseribed as separate species.
Characters of Shell—Shell of medium size (maximum length according to D’Orbigny, 70 mm.; my largest is 62 mm.), solid and rather heavy. Outline sub- trapezoidal, more or less elongated, straight (not oblique), but very variable. The upper margin may be rather straight (chiefly in young ones), or more or less curved (in older ones), with or without a distinct posterior upper angle. The posterior end of the shell is more or less pointed; the position of the point is variable, but generally rather low, and little elevated above the base-line. The lower margin is gently curved, often nearly straight in part (chiefly so in old shells), and never curved up suddenly in its posterior part, but only gently and gradually so, if at all. The posterior end of the shell is thus distinctly more tapering than the rounded anterior end, the posterior point lying rather low. The proportion of height to length of the shell is very variable, ranging from 48 to 65 pr. ct:
Valves quite convex and swollen, hardly flattened upon the sides, but more convex anteriorly and over the posterior ridge, which is blunt, but more distinct towards the beaks (and in young shells). Diameter 34 to 50 pr. ct. of length. Beaks a little inflated, but not very prominent, located at 21 to 29 pr. ct. of length. Beak-seulpture seen only in my youngest specimens, extending not more than 8 mm. from the point of the beak, consisting of about thirteen radial bars, of which only the lower ends are seen. They increase little in length posteriorly, are rather sharp in front, but the longest are somewhat obtuse, while the two last, right upon the posterior ridge, are again sharp and shorter. In some of my specimens, chiefly the younger ones, there are some short, oblique wrinkles (one to eight) upon the posterior slope; in others they are entirely absent. Lunula absent or present, narrow.
Epidermis with numerous, irregular, finer and coarser, concentric lines; the finer ones sublamellar on posterior slope and towards the margin. Radial seulp- ture obscure, but present in the shape of fine lines, sometimes more distinct on the anterior portion of the shell. Color of epidermis dark olive-green to black.
“ Von Ihering (1893, p. 102) reports athiops from the Guahyba drainage in Rio Grande do Sul,
and a variety (piracicabana) from the Upper Parand-drainage in Sao Paulo, but these records should be doubted.
508 MEMOIRS OF THE CARNEGIE MUSEUM.
Old specimens are generally uniformly black; younger ones are dark greenish olive or brownish olive, sometimes with more or less distinct concentric lighter (brownish to yellowish) bands.
Hinge-line gently or more strongly convex. Ligamental sinus over the pos- terior part of the lateral teeth; in larger specimens over the last fifth. Laterals rather straight in young, curved in older shells, one in right, two in left valve, their edges somewhat rough. Pseudocardinals normally two in right, and one in left valve, but often there is a second (posterior) one in the left valve. These teeth are extremely variable. In younger shells they are compressed and lamellar, directed obliquely forwards; the posterior tooth of the right valve is always thicker and higher than the anterior, and they are, chiefly the former, crenulated or denti- culated. In older shells, the posterior tooth of the right, and the (anterior) tooth of the left valve may become thicker, more stumpy, and may be much divided. The posterior tooth of the left valve, if present at all, may be larger or smaller, compressed or stumpy. .
Cavity of shell and beaks moderate. Nacre in all my specimens white, iri- descent posteriorly. Anterior adductor-scar distinct and impressed; anterior retractor-sear separated from it, small and deep; anterior protractor-scar connected with it. Posterior adductor-sear less impressed, the posterior retractor-sear forming an upper process of it. Pallial line distinct. Dorsal muscle-sears a few, situated in the beak-cavity.
MEASUREMENTS.
No. | Sex. | Length. Height Diameter. Beaks. 14. | og | 34 mm. | 20.5 mm. =60 pr. ct. of L. 12 mm. =35 pr. ct. of L. at 10 mm. =29 pr. ct. of L. 12. fou | 48 os 26.5 =55 3g | 20) oA is 10 SrA = ce rol 54 Hs 35 =65 a 2h 30) Me 14 =26 . oe 2 2 54 is | 30.5 =56 | 24 4 rs 13 =24 : Half shell... . . .| 58 Gs | 3l =53 + ee One — 3) 3 13 OC =D ~ 3.. Q | 61 HalesD =57 S ek A Se xs 17.5 ‘ =29 5 2 | 62 ‘= | 35 =56 2h SS =44 si | 15 “ =24 -
For comparison I give here previous measurements. Length Height. | Diameter. Beaks.
charruanus, D’Or-
bigny’s text: 70 mm. 61 pr. ct. of L. | 45 pr. et. of L. rhuacoicus, D’Or-
bigny’s text: | 63° mm. 51 pr. ct. of L. 43 pr. et. of L. faba, (rhuacoicus), |
D’Orbigny’s fig. 43 Se 20.5 mm. =48 ee 15 mm. =35 = 12 mm. =28 pr. ct. of L. athiops, Lea's fig. | 53 2 28 oe 18 =34 2 “es bs parcus, Lea’s fig. 36 “ 18 >i) 13 =36 10 =28 4 hidalgoi, Haas: 63 *e 39 “« =62 ai, 43
66.5 42 =63 2 AL
fortis, Marshall: 66 a y 37 .
ORTMANN: SOUTH AMERICAN NAIADES. 509
Remarks.
Among our specimens the single valve comes nearest to the measure- ments given for U. rhuacoicus. D’Orbigny himself admits that this may be only a more elongated form of charruanus, and his faba, united by him with charruanus, is even more elongated. There is no reason for keeping rhuacoicus separate, since the proportional length is said to be the only difference.
The variations in the shape of the shell are, indeed, very great. Younger specimens are, however, more uniform; they are more elongated, and their normal shape is well rendered in D’Orbigny’s figures of faba (Pl. 71, figs. 12-14), and in the figures of wthiops and parcus. The latter undoubtedly is a young shell, but ethiops also belongs here. Specimens like those measured under Nos. 12 and 9 resemble this very much, except that they are a little more swollen, and our smallest (No. 14) looks very much like parcus, but it is less elongated.
In the description and the figures of D’Orbigny only the color of the epidermis is not exactly as in our specimens. D’Orbigny describes it in charruanus as well as rhuacoicus as brownish green, and ‘figures it as lighter or darker brown, while the color of faba (Pl. 71, figs. 13, 14) is blackish, agreeing better with our specimens. However, our young specimens have a color, which may be called brownish green, or rather ‘“‘olive-brown.” Lea’s wthiops is said to be black.
The chief characters of this species thus seem to be the subtrapezoidal, rather elongate, straight shell, with a moderately sharp posterior point, located only little above the base-line; the somewhat swollen shell, with a moderate and blunt posterior ridge, and the brown to blackish color of the epidermis. The chief varia- tions are found in the length of the shell, and the somewhat higher or lower position of the posterior end.
There is no question that hidalgoi Haas and fortis Marshall are this species. The former is founded upon two specimens, the latter upon a single one, which certainly represent individual phases. Specimens greatly resembling these are among my material.
Anatomy.—I have the soft parts of eleven males and six gravid females, three with eggs, three with glochidia, one of these discharging.
Anal opening slit-like, closed above, the closed part about four times as long as the opening; the latter shorter than the branchial opening. The two openings are separated by a solid mantle-connection. Branchial with small papillze. Palpi subtriangular, their posterior margins connected near the base. Gills posteriorly of about the same width, but anteriorly the inner is wider, and its anterior end is immediately behind the palpi. Inner lamina of inner gill entirely connected with abdominal sac. Non-marsupial gills with scattered, short, interlaminar connec-
510 MEMOIRS OF THE CARNEGIE MUSEUM.
tions. The marsupial part of the female occupies a large portion of the inner gill, leaving about one-fourth of the gill free in front, and a smaller part free behind, thus gravitating slightly toward the posterior part of the gill. The interlaminar connections could not be distinetly observed by me on a lateral view, since no barren females are at hand; but from sections it was possible to infer that they form in- terrupted septa forming a system of intercommunicating water-tubes. The eggs and glochidia fill the water-tubes and the perforations of the septa in a mass, which is not conglutinated and divided into placente.
The glochidium (Text-fig. 4g, p. 469) has the characteristic triangular shape, somewhat oblique, with the point situated below the posterior end of the upper margin (like the figure of the glochidium of U. peculiaris, See Lea, Obs. XII, 1869, Pl. 34, fig. 80). This point does not possess a hook. Size of glochidium: L. 0.31, D. 0.26 mm.
I have examined the glochidia of three specimens; one of these had the mar- supium largely empty, and thus it appears to have been discharging. Yet no hooks were seen. But it may be that the discharge in this case was premature, that none of the glochidia were mature, and that the hooks might have developed later. This can be decided only by investigating more material.
15. DreLopon picreus (Lea) (1860). Anatomy of gills: Pl. XLVI, fig. 2; glochidium: Text-fig. 4h, p. 469. Unio piceus Lna, Obs., X, 1863, Pl. 41, fig. 287. Diplodon piceus Simpson, 1914, p. 1244; Haas, 1916, pp. 15, 49.
Type-locality.— Uruguay River.
Other Localities—Rio Uruguay, Salto Oriental, Uruguay (Haas); Rio Migue- lete, Uruguay (Haas).
Localities Represented in the Carnegie Museum.—Rio Uruguay (in mud), Uruguayana, Rio Grande do Sul, Brazil (J. D. Haseman coll., February 5, 1909) eight specimens, seven of them with soft parts, males and gravid females. Arroyo Miguelete, Montevideo, Uruguay (J. D. Haseman coll., February 17,1909) one specimen.
Distribution.—Positively known from the Uruguay River, and from a small coastal stream (R. Miguelete) near Montevideo, and probably more widely dis- tributed in the ‘‘ Banda Oriental” in Uruguay. It possibly may be only a form of charruanus. Corsi does not mention it from Uruguay.
A species closely allied to D. charruwanus, and very near to it in its dimensions, except that it does not show the same extremes of variation. It is, however,
ORTMANN: SOUTH AMERICAN NATADES. 511
shorter on the average (H. 59 to 63 pr. et. of L., while D. charruanus varies from 48 to 65 pr. ct.). In addition D. piceus is not so subtrapezoidal in outline, but rather subovate, which is brought about by a stronger curve of the lower margin, which ascends much more distinctly posteriorly, so that the posterior end of the shell is more elevated above the base-line. At the same time the posterior end is rather blunt and rounded, and not so pointed as in D. charruanus.
All other characters of the shell are similar to D. charruanus. The radial sculpture of the epidermis is more distinct. The inside of the shell is also similar, but the pseudocardinals are simpler, always elongate and compressed, and there is always only one in the left valve. They are not much cut up, but only crenulated and rugose. Nacre white, though one of my specimens has purplish blue in the cavity, probably a discoloration.
Beak-sculpture, as far as can be seen in smaller shells, similar to that of D. charruanus. But there is a very small specimen, which is only 10 mm. long, among them, which is doubtfully referred here. In this ease the sculpture consists of fifteen radial bars, of which the eighth and ninth unite at their lower ends. The anterior bars are shorter, but there is not much difference in this respect from the posterior bars. The bars are rather sharp and fine, but those upon the umbonal ridge are slightly thicker, and the two last, upon the posterior slope are very fine and shorter.
MEASUREMENTS.
No. | Sex. | Length. Height. | Diameter. Beaks. F |
Mode 2 es 2 | 30 mm. 19 mm. +63 pr. ct. of L. 11.5 mm. =38 pr. et. of L. at 8.5 mm. =28 pr. ct. of L. pen oe | 2 | 35 #3 PAU RSs 53!) MS 13 aoe rs 10 29 3 Bie uel ro ill acy! Ye 31 ae Or ~ 21 AL " 15 Se =29 Bteret hes 2) || era ES 30 os * | 20.5 =40 wf 15 =) =29 CRIS ote Ore GAs son! Ee Ate) “ 28 se ‘ Cos Montevideo. 57 * SOF 25.5 =45 16 FS Lea’s fig. | 47 * 28 50), ay | 19 10 12 ey “
Remarks.—There is not the slightest doubt that my specimens represent the U. piceus of Lea, but the question of its possible identity with D. charruanus must be left undecided, also that of other possible synonyms (lepidus Lea, and firmus Lea). I cannot unite these for the present on account of certain peculiar structures in the anatomy, which will be pointed out presently.
Anatomy.—I have the soft parts of one male, one barren female, and five gravid females, three of the latter with eggs, and two with glochidia, but in one only are the glochidia mature.
The structure is entirely like that of D. charruanus. With regard to the mar-
512 MEMOIRS OF THE CARNEGIE MUSEUM.
supium (Pl. XLVI, fig. 2) it is to be remarked that it occupies in larger specimens a somewhat greater portion of the inner gill, leaving anteriorly as well as posteriorly less than one-fourth of the gill free. In smaller females the marsupial part is relatively smaller. The interlaminar connections form distinct; but interrupted, septa. The connections stand very close, and the septiform structure prevails throughout the marsupium, and very little is seen of a transverse or reticulate arrangement. In some parts, chiefly towards the margin, the septa become more or less continuous. Thus there are here rather distinct, but intereommunicating water-tubes (ovisacs), which are filled, when charged, with masses of eggs connected with each other through the communications, so that no placenta-like arrangement is observed.
Glochidium (Text-fig. 4h, p. 469).—Higher and more upright than that of D. charruanus; L. and H. about the same, 0.28 to 0.29 mm. Thus it is more like that of D. firmus figured by Lea (Obs., XII, 1869, Pl. 34, fig. 82), but not quite as upright. At the point of the lower margin there is a hook of the usual shape, about 0.09 mm. long. But such hooks are present in only one of my specimens, in the other the glochidia are too immature.
It would be quite remarkable if in two species, so closely allied as D. charruanus and piceus, the glochidia should differ so fundamentally, that in one there are hooks, and in the other not. But judging from my material, this is the case. How- ever, it must be emphasized again that my material is scanty, and possibly in the ease of D. charruanus I do not at all have ripe glochidia.
16. DrpLODON URUGUAYENSIS (Lea) (1860).
Unio uruguayensis Lma, Obs., X, 1868, Pl. 45, fig. 298. Diplodon wruguayensis Simpson, 1914, p. 1234. Unio apprimus LEA (1866), Obs., XII, 1869, Pl. 33, fig. 78.
Type-locality— Uruguay River.
New Locality.—Pond along Rio Negro, Santa Isabel, Uruguay (J. D. Haseman coll., February 11, 1909). Five complete specimens, and several odd valves, two specimens (male and female) with soft parts.
Distribution.—Iknown only from Rio Uruguay and Rio Negro.
The synonymy and affinity of this species is obscure. Simpson thinks that it is close to D. wymani, but the latter is a compressed shell, while wruguayensis is much swollen. U. apprimus also is a swollen shell. Lea has already suggested that this is close to uruguayensis, but that it differs chiefly in the hinge-teeth,
ORTMANN: SOUTH AMERICAN NAIADES. 513
which are more lamellar and compressed in the latter (a smaller shell), and more stumpy and cut up in apprimus. My material shows that the character of the hinge-teeth changes with age. In general my specimens correspond in size and shape to uruguayensis, but the larger ones have more stumpy teeth, and thus I believe that apprimus is an old specimen of uruguayensis. Both Simpson and Haas (1916, p. 12, 47) unite apprimus with wymani, which cannot be correct on account of the difference in obesity. I think wymani belongs to lacteolus (See below).
However, it is quite possible that all these forms are variations of one and the same species, and then, of course, our general arrangement must be changed. As regards the present form I can only say that it looks like a very large and heavy charruanus, the shell being rather elongated, subtrapezoidal, and much swollen. In lacteolus, the shell is higher and more ovate, and much more compressed. In none of my specimens is the beak-sculpture seen.
M®8ASUREMENTS.
No. Sex. | Length. Height. | Diameter. Beaks. Ostepereiete | 73° mm. | 50 mm. =68 pr. ct. of L. 31 mm. =43 pr. ct. of L. at 19 mm. =26 pr. ct. of L. eo ees ee f WS.) >: 47 et iy! a 37.“ =50 s 2 Oe = Dee ae ache | 52“ =67 Aes ol Die =2i, VE he hte Oe gs: My | 52 Cds oa ao ae = 19, “ =24 P (Gee sored 79 aia 60 Me Sarasa ms 20 oe) uruguayensis..... 70 | 45 4. ¥ |} 30 “ =43 me apprimus'®....... 101 66 “ =65 re | 40 “ =40
Anatomy.—Soft parts of a male and a barren female at hand.
Color of soft parts: distal part of foot dark gray, this color sharply marked off from the whitish basal part.
Anal closed above; closed part about four times as long as the open part, which is slit-like, and shorter than the branchial, and separated from it by a solid mantle-connection. Branchial opening with small, but distinet, papillae. Palpi rather large, subtriangular, with lower margins strongly convex; posterior margins connected at base.
Gills long and moderately wide, the inner one much wider than the outer anteriorly, its anterior end immediately behind the palpi. Inner lamina of inner gill entirely connected with abdominal sac. Non-marsupial gills with scattered, short, interlaminar connections. In the female a large section of the inner gill is marsupial, about one-fourth of the gill remaining non-marsupial at the anterior
18 The measurements of uruguayensis and apprimus are those given by Simpson for the respective types. Those for apprimus are given by him under wymani (p. 1231).
514 MEMOIRS OF THE CARNEGIE MUSEUM.
end, and less than that at the posterior end. The interlaminar connections of the marsupium are developed as interrupted septa, forming communicating water- tubes. The connections are short, and are also arranged in irregular cross-rows; but towards the edge of the gill, the septiform arrangement is quite distinct.
17. DreLopon HILDA Ortmann, sp. nov.
Shell: Pl. XXXVI, fig. 7; Pl. XX XVII, figs. 1, 2, 3; Anatomy of gills: Pl. XLVI, fig. 3; Glochidium: Text-fig. 47, p. 469.
Type-locality—Rio Jacuhy, Cachoeira, Rio Grande do Sul, Brazil (J. D. Haseman coll., January 26, 1909). T'ype-set: Carnegie Museum, No.. 61.5864. Fourteen specimens, males, barren and gravid females, soft parts of all in aleohol.
There were six additional specimens in the original lot.
Description of the Shell.—Shell rather small (maximum length 46 mm.), rather solid. Outline subelliptical or indistinetly subtrapezoidal, height 57 to 63 pr. ct. of length. Upper margin straight or gently convex, forming a blunt angle with the posterior margin. Posterior margin obliquely descending, straight, or gently convex, forming a blunt point with the lower margin; this point situated distinetly above the base-line. Lower margin rather regularly convex, ascending anteriorly and posteriorly, its lowest point situated between beaks and posterior end of ligament (rather median). Anteriorly the margin is regularly rounded. Anterior end of shell not much narrower than the posterior, which tapers to the blunt pos- terior point, so that the shell is rather regularly elliptical, with the posterior end subpointed, the angle of the posterior upper margin giving a suggestion of the subtrapezoidal shape.
Valves rather regularly convex, greatest diameter slightly behind the middle, but distinetly in front of the posterior ridge, which is very blunt and broad. Pos- terior slope somewhat compressed, sometimes with a trace of a radial furrow. Sides of disk less convex, but not at all flattened. Diameter 34 to 44 DE Gt: Of length, so that the shell appears as moderately swollen. Beaks a little inflated, but not very prominent above the hinge-line, located at 23 to 28 pr. et. of the length. Beak-sculpture seen only in the smallest specimens, and only the lower part of it. There are about fourteen radial bars, the anterior rather sharp, the posterior located just in front of the posterior ridge, broader (at least their lower ends) and the latter are a little longer than the former. The last two or three bars are again finer and shorter. The longest bars are about 8 mm. long. There may be fine, short, oblique wrinkles upon the posterior slope, but these are distinet only in a few specimens. Lunula short, very narrow, or practically absent.
ORTMANN: SOUTH AMERICAN NAIADES. 515
Npidermis shining, chiefly so in the middle of the disk and in younger speci- mens, but with unequal concentric wrinkles and striaw, which become somewhat sublamellar on the posterior slope and near the margins. Radial sculpture present, consisting of irregular, often Beuiem ated lines, chiefly upon the anterior part of the shell, often appearing as “scalariform stripes’’ (radial rows of fine, short, con- centric wrinkles). Color i epidermis light to dark brown; in young specimens it is a beautiful, shining, golden brown, lighter in the middle of the shell and towards the beaks (and sometimes here with light greenish shades). In older shells the color is chestnut-brown to dark brown. In some specimens there are indications of darker concentrie bands. No traces of color-rays, except one or two very faint dark rays upon the posterior slope (seen only when held up against a strong light).
Hinge-line very gently curved. Ligamental sinus over the posterior fourth or third of the lateral teeth (more posterior in old shells, which shows that the ligament is comparatively longer in them). Lateral teeth gently curved, long, one in right, two in left valve, finely rugose. Pseudocardinals obliquely direeted forward and downward, normally two in right, one in left valve, compressed, but not very long, their edges crenulated and serrated. The posterior in the right valve more elevated than the anterior, and often thicker. Sometimes there is a trace of a second posterior tooth in the left valve, but this is always small.
Cavity of shell and beaks moderately deep. Nacre whitish, shining g, in the largest specimens the thickest parts (toward the front of the shell) have a faint rosy hue. Musele-scars of the typical shape; the anterior adductor-scar is rather deep, the retractor-scar separated from it, round and deep, the protractor-sear connected with it; the posterior adductor- and retractor-sears are united. Pallial line distinct. An irregular, longitudinal row of dorsal scars in the beak-cavity.
M®ASUREMENTS.
No. Sex.| Length. m | Height. | Diameter. Beaks. Figure a.
(hes Z ae mm. 114.5 mm. =58 pr. et. of L,| $8.5 mm. =24 pr. ct. . of Ib lat 28 pr. ct. of L.| Pl. XX XVII, fig. 1. (ope © /19.5. Ot) i eb“ s==S5, Us re
Rests : 23.0 ““ =63 . Spee =A i
p...| 2 | ane “124 aa 00) 5 16.5 “ =44 ;: ?
q a | 43 ie 24.5 CO Sei - \17 *=40 ye Pl. XXXVI, fig. 7. UBeeriee: le 40ro. eee 27. was Bs l19 AT at e Pl. XXXVII, fig. 2.
Remarks.—I have been unable to recognize this species in any of the published descriptions. It is a rather beautiful shell; marked by its small size, ne: arly regularly elliptical outline, with no obliquity, rather swollen valves, and with a peculiar, shining, golden brown color when young. It resembles, to a degree, D. charruanus, but differs from it chiefly in size, more regular shape, and the gloss of the epidermis.
o —_ or)
MEMOIRS OF THE CARNEGIE MUSEUM.
Very few species of Diplodon are known in Rio Grande do Sul, from the drainage of the Guahyba (to which the Rio Jacuhy belongs)and all have been only incident- ally mentioned by Von Ihering (1893). Under U. ethiops piracicabana (I. ¢., p. 102) he reports that U. e@thiops is found in the Guahyba River. I have been unable to discover this species, which is identical with charrwanus, in the material col- lected by Haseman in this system, and I am sure that the present form is not the one which Von Ihering calls wthiops. From his casual remarks it is seen that the latter has a shallow depression on the disk (J. ¢., p. 104), and nothing of the kind is seen In D. hilde.
Anatomy.—I have two barren females, and three gravid females with glochidia. Of the latter, one had only very few glochidia in the marsupium, but some were in the suprabranchial canals, and thus it was evidently discharging. The rest of my specimens are males.
Color of soft parts whitish; distal part of foot grayish black.
Anal opening closed above; closed part about four times as long as the open part, the latter slit-like, slightly shorter than the branchial; the latter has small, but distinet papille. Anal and branchial openings separated by a solid mantle- connection. Palpi moderate, subtriangular, posterior margins connected at base.
Gills long and rather wide. They are about equally wide posteriorly, but the outer is subtriangular and narrows anteriorly, while the inner does not, and re- mains as wide as posteriorly, with its anterior end immediately behind the palpi. Inner lamina of inner gill entirely connected with abdominal sac. The non- marsupial gills with scattered interlaminar connections. In the female, the mar- supium (Plate XLVI, fig. 3) is located in the inner gill, but anteriorly about one- third of the gill remains non-marsupial, posteriorly much less, about one-fifth of it or less, so that the marsupial part is located distinctly more backward in the gill, occupying about half of its length. When charged, the marsupium forms a slightly swollen patch. In young specimens the marsupium is smaller. Inter- laminar connections in the marsupium forming septiform rows, with a tendency to fall also into irregular transverse rows. The transverse arrangement prevails near the base and in the middle of the gill, here and there with a suggestion of a quincuncial disposition; the septiform arrangement is found toward the margin of the gill.
Glochidium (‘Text-fig. 47, p. 469) subtriangular, longer than high, with a ventral point, slightly oblique, the point being vertically under the posterior end of the upper margin. There are no hooks. In one of my specimens, the glochidia are not margined (not mature); in the two others they are margined, with a narrow
ORTMANN: SOUTH AMERICAN NAIADES. S17
rim around the anterior-lower-posterior margin, representing, apparently, the first rudiments of the postembryonal shell. Size (without rim): L. 0.29 to 0.30; H. 0.26 mm.; (with rim): L. 0.34 to 0.35; H. 0.28 to 0.29 mm.
The hookless glochidium, provided with a rim or margin, is highly interesting in view of the fact, that this structure has not been observed in other species of the charruanus-group.
5. Group oF Diplodon lacteolus.
Like the fourth group (that of charrwanus), but shell higher and shorter, subtrapezoidal to ovate, chiefly so when young, compressed or somewhat swollen. Beak-sculpture fine or a little heavier and better developed, but not covering a large part of the shell.
This group stands close to that of D. charruanus. The greater height of the shell is chiefly evident in the young shell, which may be slightly oblique. But the older shells are also shorter and higher than in the species of the charruanus- group, although they generally are less elevated than younger shells. They are not oblique, and have a rather regular, broadly ovate, or subelliptical outline.
18. DipLopon BURROUGHIANUS (Lea) (1834). Anatomy of gills: Plate XLVI, fig. 4.
Umio burroughianus Lea, Obs., I, 1834, Pl. 10, fig. 27; D’OrBriany, 1843, p. 609;
Von Martens, 1894, p. 164; Corsi, 1901, p. 450. :
Diplodon burroughianus Stimpson, 1914, p. 1271.
Type-locality—Rio Parana, Province of Corrientes, Argentina.
Other localities.—Small rivers of the Banda Oriental in Uruguay (D’Orbigny); Montevideo (D’Orbigny); swamps along the Parana from Buenos Aires to above Corrientes (D’Orbigny); Paraguay (Von Martens).
It also occurs near Santa Cruz de la Sierra in Bolivia (D’Orbigny) which is in the drainage of the Amazons.
New Locality —Pond near the Rio Negro, Santa Isabel, Uruguay (J. D. Hase- man coll., February 11, 1909). Three complete specimens, two of them, male and female, with soft parts, and four odd valves.
Distribution.— Drainage of Rio de la Plata, including the small streams of the Banda Oriental, Rio Negro, and the Parana and Paraguay to Paraguay, and possibly also in the Amazon-drainage in Bolivia. The latter part of the range, however, should be confirmed.
Some of my specimens, chiefly the female with soft parts, agree very well with
518 MEMOIRS OF THE CARNEGIE MUSEUM.
burroughianus in shape, dimensions, and color. Others differ more or less, chiefly in having the posterior point of the shell less elevated above the base-line. Thus they approach other species described by Lea as piger (1860) and ampullaceus (1869), both from the Uruguay River (See also ampullaceus, Haas, 1916, pp. 11, 47). My material is not sufficient to work out the synonymy.
MEASUREMENTS.
Sex. Length. | Height. | Diameter. | Beaks.
.e) 54 mm-| 38 mm.=70 pr. ct. of L. | 23.5 mm. =44 pr. ct. of L.) at 18 mm. = 24 pr. ct. of L. fot DG:0) Ja OOM ue Ihe : Lie a — 9, as LO ey Hoth 60 “ | 44 = {33 st 26 a 43 a 16 “ =27 “(Lea’s figure) 100 ee 62 LG 44 LG (D’Orbigny) 94 OD |G 4e aos UL | 40 = A3 E (Von Martens) S85 2h N58 okeee—nb S (Bomeeo—A0 xt (Simpson) eae SS Ube Si . 263i a Do.
The two specimens measured by Simpson do not include the type. They are more compressed than any others.
Anatomy.—The soft parts of a male and a female are at hand.
Color of soft parts whitish, distal part of foot grayish.
Anal opening slit-like, closed above, short, shorter than the branchial opening, separated from the latter by a connection of the mantle-margins. Branchial opening short, its inner edge with small, but distinet papillae. Palpi subtriangular, moderately large, lower margins convex, posterior margins united at base for a short distance.
Gills of the usual shape; the inner the wider, chiefly in front, its anterior end close behind the palpi. Structure of non-marsupial gills as usual. Marsupial portion in the female (Pl. XLVI, fig. 4) located in the inner gill, leaving about one- third at the anterior end, and about one-fifth at the posterior end free, so that the marsupium distinctly gravitates toward the posterior part of the gill. Interlaminar connections arranged in interrupted septa, and irregular, transverse rows, here and there quincuncial. This irregularly reticulate structure prevails throughout the marsupium, and the septiform arrangement is obscure.
19. DipLopoNn LAcTEOLUS (Lea) (1834).
See: Diplodon lacteolus (Lea) and D. wymani (Lea) (1860) Stmpson, 1914, p. 1226, 1230.!°
19 Lea himself (1834, p. 90) has identified the type of Lamarck’s U. delodonta (1819) with his lacteolus
(1834, p. 40), but not until after the latter name had been published in a satisfactory way, while the
original description of delodonta is absolutely unidentifiable. Thus, according to the rules, lacteolus
must be used, as Simpson has done.
ORTMANN: SOUTH AMERICAN NAIADES. 519
The following references should be added to those given by Simpson.
Unio delodonta D’OrsBiGNy, 18438, p. 605; Corsi, 1901, p. 449. Diplodon wymani Haas, 1916, pp. 12, 47.
As to the synonymy compare Simpson. His D. wymani only in part belongs here. D. apprimus Lea, united by him with wymani, is different (more swollen, see above p. 512). On the other hand, I cannot distinguish the real wymani from lacteolus, and Von Thering (1893, p. 117) also unites them.
Type-locality.—Rio de la Plata.
Other Localities.—Stream, Villa del Cerro, Montevideo, Uruguay (D’Orbigny) ; Rio Uruguay (Lea, wymani); Rio Uruguay, Las dos Hermanas Islands, Uruguay (D’Orbigny); Rio Uruguay, Salto Oriental, Uruguay (Haas); Rio San Salvador (trib. to Uruguay), Soriano, Uruguay (Corsi); Buenos Aires (D’Orbigny); Rio Batel and Rio Corrientes, Province of Corrientes, Argentina (D’Orbigny).
New Localities—Pond near Rio Negro, Santa Isabel, Uruguay (J. D. Haseman coll., February 11, 1909). One specimen, young. Brooklet, two miles north of the City of La Plata, Argentina (Dr. W. J. Holland coll., September 24, 1912). Two specimens.
Distribution.—Lower La Plata and Parana systems, upward to the province of Corrientes; also in the lower Uruguay and Rio Negro, and small tributaries of the La Plata in Argentina and Uruguay.
My specimens from La Plata are typical. The young specimen from Santa Isabel is interesting in having a rather high shell, which is slightly oblique, but comparison with the other specimens shows that, according to the growth-rests, the latter had the same shape when young. This young shell also exhibits the comparatively heavy beak-sculpture, with the radial bars running down the shell about 13 to 15 mm. Traces of these are also seen in my larger specimens.
MEASUREMENTS.
Length. Height. Diameter. Beaks.
Santas belcmery.r.0.- 0 crests) ete | 50 mm. |36.5 mm. =73 pr. ct. of L.| 18 mm. =36 pr. ct. of L. |at 18 mm.=26 pr. ct. of L. abPlata Myre iey seat noe jes TS By Ro ee Z 32 ee= 38 “ 21 =25 a
Ost Fersterctac ani ctanars sin eee 87 “ {63 < =72 oe 34 “ =39 os 22 “* =25 - lacteolus (Lea’s figure)........)80 “* |51 ase 2 ol =39 ns 23 =29 i:
Do. (Acc. to Simpson)..... }80 “ {50 a — sy o 30 =38 se
Do. IDO, ee eS A }72 “ [42 SS Pdi HE Eats) 5
Do. DOr PPh. nee (ee °** 156 aoe) ia St delodontus (D’Orbigny)........)85 “* 70 35 wymant (Lea's figure). ........ 175 ‘ |49 =65 28 Ss : ee
The first measurements given by Simpson apparently refer to the type of lacteolus, although they do not fully agree with Lea’s figure. The second measure-
520 MEMOIRS OF THE CARNEGIE MUSEUM.
ments given by him belong to a specimen which is exceptionally low. The measure- ments given by Simpson for wymani do not belong to this species, but to apprimus (See above, p. 513, footnote 18).
I have a suspicion that D. felipponei Marshall (1917, p. 381, PI. 50, figs. 1-3, Pl. 51, fig. 1) from Maldonado and other places in Uruguay, is also this species. The height as given is 70 to 72 pr. et. of the length, and the general shape and other particulars agree, as for instance the hinge-teeth. But, according to the figure, felipponei is less pointed behind, and the diameter is a little greater (40 to 46 pr. et. The color also (yellowish-chestnut) is not exactly like lacteolus, which has chestnut-olive-green tints. Thus I leave this question undecided.
20. DiIPLODON MOGYMIRIM Ortmann, sp. nov.
Shells: Pl. XX XVII, figs. 4, 5, 6, 7; Anatomy of gills: Pl. XLVI, fig. 5; Section of gills: Pl. XLVIII, fig. 2; Glochidium: Text-fig. 4k, p. 469.
Type-locality.— Creek near Mogy Mirim, Sao Paulo, Brazil, tributary to Rio Mogy Guassti and Rio Grande, upper Parans-drainage. J. D. Haseman coll., August 28,1908. Type-set: Carn. Mus. Cat. No. 61.9260, fourteen specimens, males, barren and gravid females, all with soft parts. (The original lot contained over one hundred specimens. )
Of all the species of Diplodon known from the upper Parand-drainage in Sao Paulo only one described by Von Ihering resembles this in shape: Unio greeffeanus (Von Ihering, 1893, p. 96, Pl. 4, fig. 8). However, the color of the epidermis of the latter is described as being dark green to blackish, and this does not at all fit our specimens, which are brownish black, without any distinct greenish tints. Moreover, the dimensions given for the two specimens deseribed by Von Thering; although fallmg within the range of variation of my specimens, are rather extreme, and do not represent the normal condition of our species. The height is 61 and 64 pr. ct. of the length, while in my material this proportion varies from 62 to 74 pr. ct.; and the diameter is 34 and 33 pr. ct., while it ranges, in my specimens, from 32 to 42 pr. ct. U-. greeffeanus comes from the Piracicaba River (Campinas and Piracicaba), while our specimens belong to the Rio Grande drainage.
From the latter, and especially from Rio Mogy Guassu, at Jaboticabal (not “Taboticabal” as printed), Simpson (1914, p. 1250) has deseribed Diplodon trivialis. Dimensions and description agree with our specimens to a degree; but again the color is different, being described as black or dark brown, and tinted green. when rubbed, while in our specimens, when rubbed (cleaned), the epidermis is of a peculiar
ORTMANN: SOUTH AMERICAN NAIADES. 520
golden brown, without green. Also the epidermis is not at all “cloth like,” as described in trivialis (when fresh), and the description of the pseudocardinals of the left valve does not agree at all. There are said to be two pseudocardinals, the anterior one sometimes feeble, while in our specimens there is generally only one well-developed, and this is the anterior, and if there is a smaller second pseudo- cardinal, this is the posterior. The figures of trivialis given by Marshall (1917, p. 386, Pl. 54, figs. 5-8) show also that the outline is different, being evenly rounded behind. Among my numerous specimens there is not a single one which shows this character, and thus I cannot identify them with D. trivialis.
Description of Shell.—Of moderate size (maXimum length 68 mm.), rather solid. Outline short subelliptical or subovate, or subrhomboidal, when young. Height from 62 to 74 pr. et. of length. Upper margin nearly straight when young, more or less curved when old, in the first case forming an angle with the obliquely and rather steeply descending posterior margin, in the latter case passing into it more or less gradually. Posterior margin gently concave, straight, or gently convex, forming a rounded angle with the lower margin, which may be more dis- tinct in older shells. This posterior point is more or less elevated above the base- line. Lower margin gently and regularly curved, in older shells more nearly straight in the middle. Anterior end of shell slightly narrower than the posterior in young shells; in old shells this may be reversed. Thus the shell is, when young, more subrhomboidal, with an upper posterior angle (somewhat subalate), and, when old, the shell becomes subelliptical or subovate, with the posterior end a little tapering.
Valves rather regularly and evenly convex, sides not distinctly flattened. Greatest diameter a little anterior to the middle. Posterior ridge present, but rounded and indistinct, often (chiefly in young specimens) marked by a shallow radial groove running down the posterior slope, which thus appears as compressed and slightly elevated toward the upper-posterior margin (subalate). Diameter 32 to 42 pr. ct. of the length, so that the shell is rather compressed. Beaks not inflated and not much elevated, located at from 25 to 31 pr. ct. of the length. Beak- sculpture consisting of fine and short radial bars, hardly more than 5 mm. long, fifteen to eighteen in number, those in the middle converging at their lower ends (one or two pairs). They are hardly longer upon the posterior ridge, and not ap- preciably thicker. A few oblique wrinkles may be present upon the posterior slope. In most cases the beak-sculpture is entirely obliterated by erosion of the beaks, and in general it is fine, short, and poorly developed. Lunula present, narrow in young specimens, wider in old ones, but very variable.
522 MEMOIRS OF THE CARNEGIE MUSEUM.
Epidermis somewhat shining, with unequal, irregular, concentric wrinkles, more crowded and sublamellar upon the posterior slope and near the margins, chiefly in older shells. Radial sculpture may be present, but indistinet, visible
DO es
chiefly upon the anterior part of the shell as ‘‘scalariform”’ stripes. Color of epi- dermis from yellowish brown to dark brown and blackish, but without any dis- tinct traces of greenish tints. In young specimens, when well cleaned, the color is generally yellowish or golden brown in the middle of the disk, shading to chestnut- brown toward the ends and margins. Older shells are more uniformly chestnut- brown to blackish brown (often coated with a dull black-brown deposit). Slight traces of darker brown concentric bands are rarely present.
Hinge-line generally distinctly curved. Ligamental sinus over the posterior half or third of the laterals, generally very indistinct in old specimens. Lateral teeth curved (less so in young shells), moderately long, one in right, two in left valve, their edges and sides corrugated in old shells. Pseudocardinals directed obliquely downward and forward, compressed and lamellar, thin in young shells, thicker in old ones. The right valve has nearly always two of them, equally high, but the anterior narrower. The left valve has mostly only one; but there may be a small and short posterior one, and even a small and narrow anterior one. In one specimen (one out of over one hundred), the pseudocardinals are exactly re- versed: one in right, two in left valve, while the laterals are normal. The edges of the pseudocardinals are rugose and crenulated, but not dissected.
Cavity of shell and beaks moderate. Nacre whitish, but very generally parti- ally discolored, with irregular yellowish, brownish, or grayish spots. Anterior adductor-scar impressed, rounded or subtriangular; anterior retractor-scar separated from it, rounded or irregularly oval, impressed, but not remarkably deep; anterior protractor-scar connected with adductor-sear. Posterior adductor-scar ovate or subtriangular, less deeply impressed; posterior retractor-sear generally separated from it, but sometimes only indistinctly so. Pallial line distinct. Dorsal sears few and irregular, in beak-cavity, forming an indistinct longitudinal row.
MEASUREMENTS.
No Sex.| Length. | Height. | Diameter. Beaks. Figured. 49. | 31 mm.|22) mm.=71 pr. ct. of L.}11 mm.=35 pr. ct. of L.'at 8.5 mm. =27 pr. ct. of L.|
12 oO | 39 Ve eda (ars) My IPA ee ay > 11 $ 8 2 Pl. XXXVII, fig. 6. 18:..| 9 | 40:5 “* |30 yi! i 15 SS 3y, a 11 ; 27 ee
9. o' | 53 “136 “« =68 a 20 “« =88 Wr = IB Gy 4 as Ho Pl. XX XVII, fig. 5. ATs HONG See se 39, ee =e ae 26) eal : No aie O27 me
Sun sf 2) | 64.5. & 140 a O2) oe 27 AZ, 7 17 20) ry 34... 9 | 68 “143 = 183) * PASS SF 8453 21 ol x
1625S) 1968 Salad OD i. 26 oo s}s! “8 18 20 &
ORTMANN: SOUTH AMERICAN NAIADES. 523
Remarks.—This species somewhat resembles D. lacteolus, but is considerably smaller, has finer and shorter beak-sculpture, and, when old, has more simple, less dissected pseudocardinals, of which those of the right valve are more nearly equal. The young shell (Pl. XX XVII, fig. 6a) has rather a subrhomboidal shape, exactly as D. lacteolus, which is due to the better development of the angle between the upper and posterior margins, which appears as slightly elevated (alate). ‘Thus the young shell is slightly higher in its posterior section, with the posterior end of the shell less elevated above the base-line, giving to the whole shell a slightly oblique appearance. But this juvenile shape is sooner or later obliterated, the posterior wing disappearing, and the posterior end becoming more tapering, giving to the shell a rather subovate outline, with the posterior end subpointed; but this point is never very distinct. The outline of Von Thering’s greeffeanus (PI. 4, fig. 8) comes very near to the normal shape of the old shell of D. mogymirim. The brownish color of the epidermis of our species is also characteristic, and the complete absence of distinctly greenish tints is to be noted.
Anatomy.—I have a great number of the soft parts of males and of barren and eravid females, the latter partly with eggs, partly with glochidia in various stages of development. Of fifty specimens the anatomy has been investigated more closely. For the breeding season the date of collection (August 28) should be noted.
Color of soft parts brownish white.
Anal opening slit-like, closed above; closed part not quite twice as long as the open part. Anal about as long as the branchial, separated from it by a solid mantle-connection. Branchial opening with distinct papille. Palpi moderate, subtriangular, posterior margins connected for about one-third of their length.
Gills (Pl. XLVI, fig. 5a, b, c) rather wide, the outer one subtriangular, pos- teriorly slightly wider than the inner; the inner one subtrapezoidal, anteriorly wider than the outer, its anterior end immediately behind the palpi. Inner lamina of inner gill entirely connected with abdominal sac. Non-marsupial gills (PI. XLVI, fig. 5a) with few, seattered interlaminar connections. The marsupium (Pl. XLVI, fig. 5b) of the females located in the inner gills, but occupying only a part, anteriorly leaving free not quite one-third of the gill, and posteriorly hardly one-fourth, so that the marsupium gravitates slightly toward the posterior section of the gill. In young females, the marsupial part (Pl. XLVI, fig. 5)) is much smaller, and lies distinctly behind the middle of the gill. Structure of the mar- supium (Pl. XLVI, fig. 5¢ and Pl. XLVIII, fig. 2a, 6) quite peculiar, consisting of uninterrupted septa, forming well isolated water-tubes. An interrupted or reticu-
524 MEMOIRS OF THE CARNEGIE MUSEUM.
lated arrangement of the interlaminar connections is nowhere to be seen. Never- theless this structure must be regarded as developed out of the interrupted con- dition of the septa, since the non-marsupial gills distinctly show the latter (PI. XLVI, fig. 5c). In consequence of the development of the water-tubes, the egg- masses fill these tubes (the ovisaes) in placenta-like bodies (conglutinated); how- ever, these are not very solid and persistent.
Glochidium (text-figure 44, p. 469) subtriangular, longer than high, rather small. L. 0.29, H. 0.283 mm. They are slightly oblique, with the lower point vertically under the posterior third of the hinge-line. They have hooks of the usual shape, about 0.09 mm. long. Immature glochidia have no hooks.
21. DipLopon suavipicus (Lea) (1856). Unio suavidicus Lea, Obs., VI, 1857, Pl. 29, f. 24. Diplodon suavidicus Simpson, 1900, p. 876; 1914, p. 1240.
Type-locality.— River Amazon.
New Locality—Rio Tapajos, Santarem, Parad, Brazil (J. D. Haseman coll., December 6-12. 1909). Six specimens and three isolated right valves.
Description.—Shell small, greatest L. 28 mm., moderately solid, angularly subovate or subtrapezoidal, but little oblique, slightly narrower in front, somewhat broader and sub-pointed behind. Height 69 to 76 pr. et. of length. Valves not gaping; dorsal margin gently curved, descending posteriorly, anteriorly descending more steeply, and passing insensibly into the anterior margin, posteriorly passing in a rather distinct (indistinct only in largest specimens) but blunt angle into the posterior margin, which descends obliquely, and is straight or very gently curved. Lower margin ascending gently in its posterior part, and meeting the posterior margin in a more or less distinet, but rounded, angle, forming the posterior point of the shell, which is a little elevated above the base-line. Anterior part of lower margin longer, sloping distinetly upward. It may be almost straight, or very gently curved, curving up into the anterior margin. Thus the anterior part of the shell appears somewhat narrower than the posterior.
Valves moderately convex, slightly flatter on the sides. Umbonal ridge rather distinct, but rounded. Posterior slope compressed, produced in young specimens into a slight wing-like elevation of the posterior angle of the upper margin. In some specimens there is a bare indication of a radial rib upon the pos- terior slope. Diameter 48 to 50 pr. et. of length. Beaks not much swollen, little elevated above hinge-line, located at 22 to 29 pr. ct. of length. Beak-sculpture rather well developed, extending upon the umbonal ridge about 10 mm. or more,
ORTMANN: SOUTH AMERICAN NATADES. 525
but not quite so far upon the rest of the shell, and covering altogether about one- third or one-fourth of the shell. There are about fifteen or sixteen radial bars, of which the eighth and ninth unite, and between the latter there is another shorter pair, also united in v-shape. These bars are rather sharp, those just in front of the umbonal ridge are hardly broader. The most anterior bars are sometimes slightly granular, occasioned by the growth-lines cutting across them. There are generally close to the beaks a few additional radial bars upon the posterior slope, and below then some oblique, irregular wrinkles, sometimes crossing the former, so as to form v-shaped angles (forming the posterior system of re-entering angles of Lea). Lunula present, short, narrower or wider.
Epidermis with numerous, fine, concentric growth-lines, sublamellar on pos- terior slope, and in the larger specimens with traces of radial lines, chiefly in the front part of the shell. Color brown to blackish brown, without color-markings.
Hinge-line gently curved. Ligamental sinus over the posterior third of the laterals. Lateral teeth gently curved, thin, one in right, two in left valve. Pseudo- cardinals obliquely descending, rather long, compressed and thin, two in right valve, the posterior more elevated, with serrated edge. In the left valve also two pseudocardinals, the anterior larger, serrated, the posterior much smaller, sometimes rudimentary.
Cavity of shell and beaks moderate. Nacre whitish. Anterior adductor- sear moderately impressed. Anterior retractor-scar separated, and anterior pro- tractor-scar united with adductor-sear. Posterior scars indistinct, united. Dorsal sears in beak-cavity.
MEASUREMENTS.
No. Length. | Height. | Diameter. | Beaks. : | = 3 E : 9° . Det artes 18 mm. |/13 mm.=72 pr. ct. of Th 9 mm.=50 pr. ct. of L.at 4 mm. =22 pr. ct. of L. 72] I 4 2] ZS Oe er US: 14 “ =76 oe | 9 “ =49 Ss 5 “ =27 a 4 21 SS =71 e | 9 a . Ca — 29 § i ee ae ST) Ba IG NOS ATS GR ee a as 298 ak Ott a 27 118.5 ‘© ='69 HY 2 “ =44 ff C29 eS Cae er IRB A Olen Sos = ef [12 = 43 P: 8) = 29 : Lea’s figure. | 21 e 15 perl a oe eto ss Ce! :
Remarks. My specimens are rather uniform in shape and proportions. There is some variation in the posterior part of the ventral margin, which may ascend a little more decidedly, giving to this margin a more distinct lower projection, and placing the posterior point of the shell a little higher above the base-line. In my largest specimen, all angles (upper posterior, posterior, and that of lower margin) are more rounded. The beak-seulpture may be more or less distinet and sharp, and the bars vary somewhat in length.
526 MEMOIRS OF THE CARNEGIE MUSEUM.
Previously this species was known only from a single individual described by Lea. There is not the slightest doubt that my specimens belong here, and one of them (No. 4) is almost a replica of that of Lea. Lea believes that his shell is a young one, and it certainly is not full-grown, but, as my material shows, this species does not grow very much larger. Von Ihering (1898, p. 120) suggests that this might be the young stage of U. wheatleyanus Lea, but this cannot be the case, since the latter has a different, much heavier beak-sculpture.
I am uncertain about the systematie position of this species, but there are certain resemblances in the shape of the shell to that of the lacteolus-group. This species is also interesting because of the fact that it is one of the few forms of Diplodon found in the Amazon-drainage.
6. Group oF Diplodon ellipticus.
Shell more or less elongated, but often rather high and short; subovate or subtrapezoidal, distinctly oblique at all stages of growth, with the longest axis forming an angle with the line of the ligament. Anterior end narrower, posterior end higher, and lower margin distinetly ascending in its anterior portion. Beak- sculpture fine or coarse, more or less developed.
The chief character of this group is the obliquity of the shell, brought about by a widening of the posterior portion of the shell in the vertical direction, so that the posterior end lies rather low, and the longest dimension is not parallel or nearly parallel to the ligament, but forms a distinct angle with it. Although an obliquity is sometimes indicated in the species of the previous groups, it generally disappears with increasing age, while in the present group the obliquity is rather emphasized in older shells. It is all-important, in order to correctly Judge as to the shape of these shells, to place them in a uniform position, always with the ligament running horizontally.
The outline of the shells of this group varies a good deal, and some of them become very high in proportion to length, so that the outline appears more nearly subrotund, similar to the shape in the subgenus Cyclomya. However, in the latter the greatest height of the shell is always situated more nearly in the middle under the middle of the ligament, while in the shells of the ellipticus-group the greatest height is more posteriorly, at the posterior end of the ligament, or even beyond that.
22. DIPLODON ELLIPTICUS Spix (1827). Diplodon ellypticum (error typogr.) Sprx, 1827, Pl. 26, fig. 1-2. Unio ellipticus WAGNER, 1827, p. 33; Von InERING, 1890, p. 163, Pl. 9, figs. 8-9;
Von Ihering, 1893, p. 108.
ORTMANN: SOUTH AMERICAN NATADES. 527
Diplodon wagnerianum Simpson, 1900, p. 877; 1914, p. 1246.
Type-locality. Rio San Francisco (Wagner).
Other Localities —Rio Parahyba do Sul, Rio de Janeiro (Von Thering, 1893, p. 115); Rio Santa Maria, Espirito Santo (Von Thering, 1910, p. 134) (drainage of Rio Doce). The form from the latter locality has been named var. santanus Yon Ihering.
Rio Piracicaba, Sio Paulo, and Rio Tamanduatahy, Sao Paulo (Von Thering). The latter two localities are somewhat doubtful, since the specimens are not exactly like ellipticus. The location and drainage of the last named river is unknown to me.
In the Carnegie Museum is one specimen, labeled ‘“ Brazil’? (Holland Collec- tion). Not quite typical.
The change of the specific name ellipticus to wagnerianus is unwarranted. Spix gave the name in the plate in connection with the generic name Diplodon. Since the latter stands, the specific name also is entitled to recognition.
In spite of Von Ihering’s re-description of the type, this species is as yet poorly known, and our knowledge of it is founded chiefly upon what Von Thering has said. The species is positively known from the Rio San Francisco and the Rio Parahyba do Sul; the other localities are more or less doubtful, since the specimens deseribed from them do not fully agree with the type. Those from the Rio Santa Maria have been distinguished as a variety.
All we can gather from descriptions and figures is that D. ellipticus is a sub- elliptical or subtrapezoidal shell, of dark green to blackish color, with rather smooth surface. It has in the anterior part of the shell a shallow depression, often producing a shallow emargination in the anterior part of the ventral margin. In outline, the shell is distinctly oblique, narrower in front, higher behind. The beak-sculpture consists of simple, fine radial bars, which are rather short, and somewhat cut up on the posterior slope by irregular, oblique wrinkles. The nacre is blueish white. Pseudocardinals somewhat compressed, but not thin, crenulated, two in right, one in left valve, but the latter with an angle at its base, representing the remnant of a posterior tooth.
Our specimen from Brazil was received as ellipticus. It agrees in most of the above characters, except that it is shorter in proportion to height, with less pointed posterior end. Also the projection of the lower margin is indistinct. The beak has eleven radial bars in front of the posterior ridge, the seventh and eighth meeting in the middle. On the posterior slope there are corrugations and fine, oblique ridges. L. 32 mm.; H. 21 mm.; D. 12 mm. This specimen might very well be
528 MEMOIRS OF THE CARNEGIE MUSEUM.
a young elliplicus, and may correspond to the var. santanus of Von Thering (smaller,
projection of lower margin less distinct).
23. DIPLODON BERTH Ortmann, sp. nov. Shells: Plate XX XVIII, figs. 1, 2, 3, 4; Anatomy of gills: Plate XLVI, fig. 6.
Type-locality.—Rio Jacuhy, Cachoeira, Rio Grande do Sul, Brazil (J. D. Hase- man coll., January 26, 1909). Type-set: Carn. Mus. Cat. No. 61.5865. Sixteen specimens, all with soft parts, including males, barren and gravid females. (There were twenty-three specimens in the original lot).
Additional Locality—Rio Vaceahy Mirim, Santa Maria, Rio Grande do Sul, Brazil (J. D. Haseman ecoll., January 29. 1909). One male with soft parts.
Distribution: Guahyba drainage in southern Brazil.
Description of Shell—Shell rather small, maximum length 65 mm.; rather solid, chiefly so anteriorly, and in old shells often much thickened along the lower anterior margin. Outline subovate to subtrapezoidal, distinetly oblique, broad and rounded, or somewhat pointed behind. Height from 55 to 67 pr. et. of length. Valves not gaping. Dorsal margin straight, or gently descending posteriorly, forming a more or less distinct obtuse angle with the posterior margin. The latter obliquely descending, gently convex, and curving around into the posterior part of the lower margin, forming with the latter in young specimens an indistinct rounded angle, which, however, may become more distinct in old specimens. Lower margin in normal specimens with a distinct rounded projection, forming the lowest point of this margin, situated far back, behind the posterior end of the ligament. From this point the lower margin curves up behind to the posterior end of the shell and this part is quite short. Anteriorly the lower margin also slopes upward, and is almost straight for a considerable distance; sometimes it is even slightly concave; then it curves up into the anterior margin. Thus the shell appears con- siderably narrower anteriorly, broader (higher) posteriorly, with the greatest height situated far backward.
Valves moderately and not uniformly convex. The greatest convexity is near the anterior end and over the posterior ridge, which is broad and not sharply marked. In front of the posterior ridge the sides of the disk are distinctly and broadly flat- tened, and sometimes even slightly concave, producing the emargination of the anterior part of the lower margin. Posterior slope somewhat compressed, very rarely with a slight trace of a rib or a furrow. Greatest diameter of the shell 32 to 43 pr. et. of the length, located well behind, upon, or close in front of, the pos- terior ridge. Thus, although rather swollen in the region of the posterior ridge,
ORTMANN: SOUTH AMERICAN NAIADES. 529
the shell appears in front of it rather more compressed. Beaks not much swollen, and not very prominent, located at from 23 to 29 pr. ct. of the length. Beak- sculpture consists of twelve to fourteen radial bars, which are sharp and rather distant from each other, the ninth, tenth, and eleventh (immediately in front of and upon the posterior ridge) are longest, about 10 to 12 mm. long; the seventh and eighth may consist of two bars united in V-shape, but this cannot be seen clearly, since in all specimens the tips of the beaks are eroded. These median bars (seventh and eighth) are also slightly shorter than those in front, and distinctly shorter than those behind them; the last two or three bars are fine and somewhat shorter, and stand upon the posterior slope, becoming indistinct. In a few of my specimens there are traces of oblique wrinkles upon the posterior slope. Lunula short and narrow, distinct only in larger specimens.
Epidermis smooth, rather shining, with numerous, closely set, fine, concentric lines, and stronger and irregular concentric wrinkles. The fine lines become sub- lamellar on the posterior slope and towards the margins. A fine radial sculpture is present, chiefly in the anterior part of the shell, but it is not very evident. Color greenish black to brownish black, darkest in old shells. Young shells are more distinctly greenish, dark olive-green, shading towards the beaks to gray-green and brownish olive. There are no distinct color-bands and no distinct color-rays, except in very young specimens, where there are traces of dark green rays on the posterior slope, seen when held up against a strong light.
Hinge straight or gently curved. Ligamental sinus over the posterior third or fourth of the laterals (more posteriorly in old shells). Lateral teeth curved, more strongly so in their posterior part in old shells, one in right, two in left valve, edges rugose. Pseudocardinals two in right, one in left valve, subcompressed, not very long, in old specimens sometimes almost stumpy. The posterior pseudo- cardinal of right valve stronger and more elevated than the anterior, often much divided, and always much crenulated. Very often there is a second posterior small pseudocardinal in the left valve.
Cavity of shell and beaks shallow. Nacre whitish, in old specimens often very thick anteriorly, and with pinkish tints; posteriorly very iridescent. Ad- ductor-scars deeply impressed anteriorly, less so posteriorly. | Anterior retractor- scar small and deep, separated from the adductor-scar; anterior protractor-scar united with it. Posterior retractor-sear connected with adductor-sear. Pallial line distinct. Dorsal sears in an oblique irregular row in the beak-cavity.
Remarks.—This species may be what Von Thering (1893, p. 102) calls wthiops in the Guahyba drainage, but only the remark (p. 104) that this @wthiops has a
530 MEMOIRS OF THE CARNEGIE MUSEUM.
broad, shallow furrow in the anterior part of the shell, seems to confirm this as- sumption. In other respects it is impossible to decide, whether this is, or is not, the wethiops of Von Ihering. It surely is not the ethiops of Lea. Of the few other species incidentally mentioned by Von Ihering as found in the Guahyba-drainage, none can be compared with our species. Its chief characters are the subovate to subtrapezoidal distinctly oblique shape, narrow in front, broader behind, and the peculiar compression of the shell in the anterior part. The beak-sculpture and color of the epidermis are also characteristic.
MEASUREMENTS.
No ‘Sex| Length Height Diameter. Beaks. | Figured.
Tague | pr. ct. of L.|Pl. XX XVIII, fig. 2.
3...| o' | 31.5 mm./21 mm. =67 pr. ct. of L.|10 mm. =32 pr. ct. of L.Jat 8 mm. =25
11. 4825) oo 27 pee O5: ok 1s Schl L2D is oe 29
14. 9 | 47 oe [olLOMinet—O7, 19 nx oA () IDs Ce —P7/
22. 2 | 51.5 “ 132 Sea O92 - 22 S85 ae 12 a
PETE | Ae EY = (29 OO) fa 210) => =A vs | 12 a2) |
24. OU] 56: “ 136 SO ae PAS BTS} x 13.5 =24 Pl. XXXVIII, fig. 1. 34. a! 65 “> (Bozo oe eo = 26 pone) 15 = 33} (Santa Maria)
I cannot compare this species with any other, except D. ellipticus Spix. The general shape is very much the same, but ellipticus seems to be more elongate (height only 54 pr. et. of length). The posterior end is slightly more pointed, and the diameter is distinctly less (31 pr. ct. on the average). In our species the average height is 62 pr. et. and the diameter 39 pr. et. In addition our shell seems to be thicker and more solid, chiefly in old specimens, where the lower anterior margin is considerably and strikingly thickened, a character not mentioned in ellipticus.
Old shells often become freakish, assuming irregular shapes. Frequently the posterior part of the shell grows more strongly in length, thus rendering the shell exceptionally long (as in No. 23). In such specimens the projection of the lower margin is obscured, and the posterior point of the shell is very little elevated above the base-line. Furthermore the shell in general is more pointed behind. Such specimens also appear more swollen. However, the growth-lines clearly indicate that, when young, these individuals had the normal shape. In other old shells the whole posterior part is more developed (No. 24, Pl. XX XVIII, fig. 1a) and is deflected downward. This fact tends to preserve the general shape, but renders the anterior part of the lower margin somewhat concave.
Anatomy.—Judging from the soft parts at hand, twelve specimens are males, five are barren females, and seven are gravid females. Only one of the latter con- tained immature glochidia. The date of collection (January 26) apparently is near the beginning of the breeding season.
ORTMANN: SOUTH AMERICAN NAIADES. 531
Color.—Distal part of foot dark brownish, grayish, or blackish; the rest of the soft parts are whitish.
Anal opening closed above; the closed part about four times as long, or a little longer than the open part; the latter slit-like, slightly shorter than the branchial opening. The latter with small, but distinct papillae, separated from the anal by a solid mantle-union. Palpi subtriangular, moderately large, of the usual shape; their posterior margins connected at base.
Gills rather long and wide. Outer gill subtriangular, widest at the beginning of the posterior third, and here it projects a little beyond the inner gill. Inner gill subtrapezoidal, its anterior end immediately behind the palpi. Inner lamina of inner gill connected with abdominal sae.
Structure of non-marsupial gills as usual. In the female, the marsupium (Pl. XLVI, fig. 6) is located in the inner gill, and in large specimens it is restricted to the middle part of the gill, leaving non-marsupial almost one-third at the anterior end, and a somewhat smaller portion at the posterior end. In young specimens the marsupium is smaller. When charged, the marsupium forms a slightly swollen, lenticular, rounded, or oblong patch in the middle of the gill, most of it lying behind the middle. The interlaminar connections of the marsupium are strongly de- veloped, forming very incomplete, interrupted septa, and arranging themselves rather in transverse and oblique rows, so that the vertical septiform structure is obscure, while a reticulate and irregularly quincuncial arrangement prevails. Only near the margin of the gill is a septiform structure indistinetly indicated. The egg-masses do not conglutinate into placente-like structures.
Only one of my females has very young glochidia. They are, as far as can be seen, of the usual shape, subtriangular, and somewhat oblique. Exact measure- ments could not be obtained. No hooks are visible, but, of course, such may be present in ripe glochidia.
24. DIPLODON ENNO Ortmann, sp. nov. Shells: Plate XX XVIII, figs. 5, 6, 7, 8; Anatomy of gills: Plate XLVI, fig. 7.
Type-locality.— Rio Grande, Boqueirao, Bahia, Brazil (S. Francisco drainage). (J. D. Haseman coll., January 9, 1908). Type-set: Carn. Mus. Cat. No. 31.9264. Eighteen specimens, males and barren females, with soft parts. (A number of additional young specimens were in the original set.)
According to the latest census of the Naiades from the Rio S. Francisco drainage (Von Ihering, 1910, p. 138), there are only two species of Diplodon present in this system: D. rotundus Spix, and D. ellipticus Spix. The former is much
532 MEMOIRS OF THE CARNEGIE MUSEUM.
higher and much more rounded than the present species; the latter is more elongated and more pointed behind, and has, besides, a smooth epidermis (Cf. Von Ihering, 1890, p. 163). Since it is impossible for me to find any other South American species of Diplodon, the description of which answers to the present species, we must regard the latter as new.
Description of Shell.—Shell small to medium (maximum length 53 mm.), rather thin. Outline subovate or subtrapezoidal, distinetly oblique, higher behind, narrowed anteriorly, the obliquity being most pronounced in older specimens. Height 56 to 75 pr. et. of length (against 54 pr. et. in ellipticus, and 86 or 87 pr. ct. in rotundus). Valves not gaping. Dorsal margin straight or gently convex, form- ing a rather distinct or obtuse angle with the posterior margin, which descends ob- liquely and rather steeply, is nearly straight or gently convex, and curves into the lower margin without forming a distinct posterior point. The posterior ex- tremity of the shell is located relatively low and is only moderately elevated above the base-line. Lower margin with its lowest point located rather posteriorly, vertically below the posterior end of the ligament, or even behind it, strongly as- cending in a curve toward the posterior end of the shell, but nearly straight or very slightly convex in its anterior part, and sloping upward toward the anterior margin, so that the anterior portion of the shell is distinetly narrower than the posterior, producing thus the oblique appearance of the whole shell.
Valves comparatively compressed. Diameter 28 to 45 pr. ect. of length. Greatest convexity and greatest diameter situated well back upon the posterior ridge, which, however, is very indistinct and broad. Sides of the shell in front of posterior ridge not very convex and rather flattened. Posterior slope compressed, sometimes with a faint trace of a radial groove, somewhat elevated (wing-like) toward the upper posterior angle. Beaks not swollen, and hardly elevated above the hinge-line, located at from 22 to 28 pr. et. of the length. Beak-sculpture sharp and fine, distinct, but restricted to the region near the beaks. There are fifteen to eighteen radial bars, the ninth and tenth in the middle, joined and v- shaped at their lower ends. . The longest bars (6 to 8 mm.) stand upon the posterior ridge, the anterior ones being distinctly shorter. There are a few anterior bars and upon the posterior slope a few posterior bars, and sometimes traces of oblique wrinkles. None of the bars are distinctly granular, and a comparatively slight degree of erosion obliterates all traces of beak-sculpture. Lunula indistinet and narrow, visible only in older shells.
Epidermis not shining, but rather rough. This is due to a great number of fine, irregular, concentric lines, which, when well-preserved, are lamellar and
ORTMANN: SOUTH AMERICAN NAIADES. 533
elevated, showing this character all over the disk, being, however, more distinctly lamellar and more crowded upon the posterior slope and near the margins. Even when these fine lamellx are worn off, the epidermis does not become shining, but remains dull, and when well-preserved, the epidermis appears cloth-like. There are no traces of radial sculpture. Color of epidermis dark greenish black, but often in the middle of the disk and towards the beaks brownish black. The greenish tint is not very evident, and is best seen in young specimens.
Hinge-line very gently curved. Ligamental sinus over the posterior third of the laterals. Lateral teeth curved, one in right, two in left valve, their edges slightly corrugated or nearly smooth. Pseudocardinals narrow and compressed, but not very long, corrugated and rugose, but not cut up, two well developed pseudocardinals of nearly equal size in right valve, one well-developed in left valve; but there often is a smaller posterior one in the left valve, and another small, low, ‘and narrow anterior one, so that the left valve may have three pseudocardinals. The middle one, however, is always the largest. Sometimes the posterior pseudo- cardinal of the right valve is higher and thicker than the anterior.
Cavity of shell and beaks shallow. Nacre blueish silvery, very iridescent, and sometimes discolored yellowish. Anterior adductor-scar distinct and moder- ately impressed. Anterior retractor-scar rounded, small and rather deep, separ- ated from adductor-sear. Anterior protractor-scar united with it. Posterior adductor-scar rather indistinct and not impressed, the posterior retractor-scar forming an upper process of it. Pallial line not very sharp. Dorsal sears few in beak-eavity.
MEASUREMENTS.
No. |Sex. Length. | Height. | Diameter. Beaks. Figured. ikea) Legal el PEGS mm.| 7 mm. =56 pr. ct. of L. 3.5 mm. =28 pr. ct. of L.'at 3.5 mm. =28 pr. et. of L. 140% 2)? |-22 Beal Wis Baye Tes Prey oe 7 se 2 | 6 Sey se
8...| o' | 28 Be TAO EES 83 7 9 =32 7 = 25
4...) | 35 ** 126 e118 12 =34 8 =23) Pl. XXXVIII, fig. 5. 2a | 45 pz = dd 16.5 =or