![Thumbnail: [Volume 68, 1938-39 page 60]](/journals/rsnz/images/rsnz_68/thumbnails/rsnz_68_01_0104_0060_ac_02.gif)
Criteria For Classifying The Pitarinæ.
To separate his three chief genera of the Pitarinæ, Callista, Amiantis, and Pitar, Jukes-Browne (1913, p. 337) gave the following criteria as mainly to be relied on:
(1) The existence in Amiantis and Pitar of a channel leading from the pit between the anterior laterals of the right valve below the anterior cardinal into the first interdental socket,
(2) the position and shape of the left posterior cardinal,
(3) the form and direction of the pallial sinus.
He further added that neither the characters of the external surface nor the bridge connexion of the anterior and posterior cardinals in the right valve are to be depended upon.
Such statements must be interpreted liberally, because small differences in the hinge may be much less significant than great differences in the sculpture. Indeed, the words “small” and “great” are themselves exceedingly loose terms when each is applied to a different character. Consequently, until a mathematical basisis accepted, it may not be possible to arrive at an accurate classification that is acceptable on all sides.
![Thumbnail: [Volume 68, 1938-39 page 61]](/journals/rsnz/images/rsnz_68/thumbnails/rsnz_68_01_0105_0061_ac_02.gif)
Further, to use any one criterion or set of criteria consistently is extremely difficult. As an example, take Jukes-Browne's (1913, p. 338) treatment of Pitar Römer. Of the genotype V. tumens Gmelin, he stated, “the left posterior cardinal resembles that of Amiantis in being confluent with the nymph along its whole course, and in this respect differs from most other species of Pitaria”; also the pallial sinus is “long, horizontal, linguiform and pointed at the anterior end like that of Amiantis callosa and A. dione.” This suggests that many of the species or groups classed in Pitar by Jukes-Browne and contributing to his conception of the genus may not really belong there. The matter is further complicated by obscurity as to the characters of the genotype itself. H. v. W. Palmer's figure (1926, Pl. 1, fig. 5), reproduced from a photo, shows a deep groove separating the left posterior cardinal for most of its length from the nymph; also Cossmann and Peyrot's (1911, p. 372) line drawing shows a definite groove the full length of the tooth. Certainly in neither of the originals could it have resembled that of Amiantis in being “confluent with the nymph along its whole course.”
N. M. Tegland's (1929, Pl. 21) reproductions of Römer's figures show a shell with a very deep escutcheon as also do Cossmann and Peyrot's figures, but the shell of Palmer's photographs has quite a shallow one. Obviously such an important genotype as V. tumens should be definitely fixed to a lectotype or neotype by someone with access to authoritative collections, and carefully detailed figures and descriptions made readily available.
Broadly speaking, the hinge is undoubtedly the best guide to Venerid classification, being generally the most conservative part of the shell; nevertheless elements of it vary to a certain extent, even within the one species.
As Jukes-Browne noted, the attitude of the left posterior cardinal in the different groups is important. In most of the Cretaceous species which have had their hinges clearly described or figured, e.g., Callistina plana (Sowby.) Europe and India, Aphrodina tippana (Conr.) North America, and Trigonocallista uzambiensis (Woods) South Africa, this tooth is very long and well separated from the nymph. The writer's description and figure (Marwick, 1924, p. 595) of Tikia wilckensi (Woods) New Zealand is wrong. The left hinge has now been further cleared of matrix and shows a long, separated tooth (Pl. 10, fig. 7). Stoliczka's figures present some doubtful cases, but certainly, in the Upper Cretaceous, the dominant hinge had a long and separated left posterior cardinal.
In the rich Eocene faunas of the Paris Basin figured by Cossmann and Pissaro there are many species in the subfamily with a free left posterior cardinal, but others such at Calpitaria have it partly fused to the nymph and some such as Microcallista have it fused throughout. In the Miocene, most of the large shells have a fused left posterior cardinal and the position is much as in recent faunas. It seems fair to infer, then, that the free tooth is a primitive characteristic and the confluent tooth has developed from it. This conclusion is supported by the ontogeny of multistriata. Juvenile
![Thumbnail: [Volume 68, 1938-39 page 62]](/journals/rsnz/images/rsnz_68/thumbnails/rsnz_68_01_0106_0062_ac_02.gif)
shells have the cardinal well separated from the nymph. (See Pl. 10, figs. 1, 2.) The character appears to be connected in some way with the shape of the shell and differs to a certain degree in different individuals within the species.
Though, naturally, great differences in the strength of the sculpture exist between different groups and to a less extent within the groups, the kind of sculpture, concentric ridges and grooves tending to obsolesce, is really very constant throughout the subfamily.
An important character that has been rather neglected in classification is the manner of setting of the ligament. In some groups, e.g., Notocallista s. str. (Pl. 10, fig. 14), the ligament-walls are quite low and have rounded edges, so that the ligament and the nymphs are well exposed on the closed shell. In others, e.g., Chionella (Pl. 10, fig. 9), the ligament-walls are relatively high, so that the ligament and nymphs are deeply sunk and little exposed on the closed shell. A typical example of this kind of ligament is developed on Lioconcha castrensis (Linné). In some Pitarinæ, although the nymphs are not deeply sunk, the walls are high through being raised as a crest along the dorsal margin (Pl. 10, fig. 3). Dall described this feature in Hysteroconcha, and a less in-curved crest is developed as a specific feature in the Pliocene and Recent New Zealand multistriata. The narrow, low ridge on the posterior inner margin of the Pitarinæ forms a useful datum by which to judge the degree of immersion of the ligament.
Most if not all of the criteria of classification show a certain range of variability in the one species, even in specimens from the same locality; for example, the set of the hinge, the shape and size of the sinus, the relative sizes of the teeth, especially the right anterior and the left posterior cardinals; the amount that the right anterior cardinal deflects the groove leading back from the anterior lateral pit, the degree of separation of the pedal retractor from the anterior adductor, and the strength and persistence of the sculpture in the middle of the disc.