Thomson. — Additions to the Knowledge of Brachiopoda. 43 



Cretaceous, and there may be a number of other stocks equally wortliy of 

 generic rank. Owing to the simplicity in external form of Terebratulids, 

 and the narrow limits of possible variation, the chances of homoeomorphv 

 are very considerable, and, unless genera are closely restricted and homoeo- 

 morphy thus excluded, zoological comparisons between the fossils of dif- 

 ferent countries cannot carry much weight for purposes of correlation. 



Since the geological record is too imperfect and the differences in ex- 

 ternal form are too slight to enable the species to be grouped in linear 

 series, thus allowing their phylogenetic relationships to be traced, it is 

 necessary, if the genera are to be closely restricted, to find some anatomical 

 characters common to groups of related species by which they may be 

 distinguished from other groups of species. The characters chosen may 

 b.^ snch as have hitherttj been considered unimportant, even in specific 

 differentia, provided that they are persistent within the group. The 

 chances are that a group thus defined wall prove to be a good genetic series. 



Buckman (1910) has taken the first step, so far as fossils are concerned, 

 towards the discovery of such characters in his treatment of the Tertiarv 

 Brachiopods from the islands of the Weddell Sea collected by the Swedish 

 Antarctic Expedition. His words are worth quoting in full : — 



" Among the material brought from the Antarctic are several specimens 

 which belong to various species of Terehratula, using that term in its wide 

 sense ; but it is probable that none of them really belongs to the typical 

 series which would be grouped around the genotype Terebratida terehratula 

 Linne sp. There are two series which differ conspicuously in the character 

 of their test. The first series shows coarse and distant punctae associated 

 with a rather thick test. . . . The second series shoAvs a finely and 

 closely punctate test . . . which is also thin, as if it were a deep-water 

 series. Ftirther, in the older specimens particularly, there is an outer layer 

 of test which is undoubtedly grooved — the grooves waved and irregular 

 . very suggestive of the ornament seen in certain species of Lower 

 Jurassic Lima (Plagiostoma). This finely punctate series is not punctate 

 so finely and minutely as Terebratida variabilis of the English Tertiary, 

 which is presumably a true Terebratida : the punctae of the Antarctic 

 species are larger and therefore seem more approximate. There is evidently 

 much yet to be learnt concerning these differences of punctuation. . . . 

 These [finely punctate] species with their thin test have much the appear- 

 ance of species of Liothyrina ; but it has not seemed desirable to place 

 them in that genus, of which the characters and limits are none too fully 

 known. Of these species the loop cannot be seen, and they show no in- 

 dication of the four internal radiating furrows which serve for the attach- 

 ment of the pallial sinuses ; and these furrows so marked in the type species 

 Liothyrina vitrea would serve in fossil species as an outward index where 

 interior details could not be seen : thus they show well through the test 

 of the Chalk species assigned to Liothyrina — namely, T. carnea, T. sub- 

 rotunda, T. semiglohosa, &c." 



There are at least two Australasian Tertiary species which agree closely 

 with those of Buckman's finely punctate species — uajnely, Terehratula tateana 

 Tenison-Woods* and Terehratula concentrica (Hutton). Both these shells 

 have a thin and finely punctate test, and T. tateana shows in addition a 



* It may be remarked that Buckman identified one of his species as " Terehratula 

 vitreoides Tate (1880) ? of Woods," a name which Tate in 1899 corrected to Teretyratuhi 

 .tateana T.-Woods. 



