﻿57 95 



have been quite overlooked hitherto; even in 1903 we find in Jordan and Starks (20, p. 71) 

 under the family diagnosis for Amphixile: "Ribs developed", and under the description of 

 A.striyalcr. "There are 11 lower ventral plates (ribs\ 2 in front of the pectoral and 9 behind". 

 And the sui)posed identity of the dorsal cuirass with inner skeletal i)arts — against which 

 however Hilgekdorf does not protest — is found again both in 1902 and 1903 in Starks (30, 

 p. 625; where the presence of ribs is also mentioned) and in Jordan and Starks (I.e. p. 70). 



2; p. 58 (20): 



Information on the inner skeleton is scarce in the earlier literature. Apart from Starks' 

 investigations on the pectoral girdle and his few remarks on some of the bones of the skull 

 (given below, see note 3 p. 96 (58)) we have information, based on personal investigations, from 

 Ci'viKR (7b), Agassiz (lb, p. 275), Günther (14a, )). 527), Heckel (15, p. 223) and Hilgendorf (17). 

 In the addition inserted by Laurili.ard in Cuvier's "Leçons" (2nd edition, Vol.1, p. 229', it is 

 stated that the first 5 vertebræ are elongated without transverse processes and that the spinous 

 processes are greatly inclined backwards, so much so that the dorsal fin projects out over 

 the tail; the number of vertebræ is stated to be 15—16 in all. Both Agassiz and Günther give 

 the correct numbei- of vertebræ, 20, but whereas A. counts 8 to the abdominal and 12 to the 

 caudal vertebræ, G. gives respectively 6 and 14. The information on the part of the spinal 

 column lying outside the cuirass is — so far as it goes — in the main correct, most complete 

 in Agassiz; on the other hand, the statements regarding the part enclosed within the cuirass 

 are very imperfect. Both have seen that the first 6 vertebræ are much elongated, but we get 

 no information regarding the connection with the armour, the relation of the interspinous 

 bones etc. Agassiz has clearly not been able to distinguish between the spinous processes 

 and the interspinous bones in this region, as he only mentions the interspinous bones which 

 lie behind the dorsal .spine; his words are: 



"Les corps des six premières [vertèbres] sont tellement allongées, qu'ils forment à eux 

 seuls toute la portion de la colonne qui est recouverte par la carapace. Leurs apophyses 

 épineuses sont filiformes et démesurément longues, surtout les antérieures, qui se prolongent 

 jusqu'à l'extrémité du tronc, ou plutôt jusque sous la grosse épine qui termine la carapace, 

 en avant de l'insertion des osselets interapophysaires qui portent les rayons épineux ou la 

 partie antérieure de la dorsale. Les deux dernières vertèbres abdominales sont courtes, 

 semblables à celles de la portion caudale; celles-ci, au nombre de douze, ne forment pas, 

 dans leur ensemble, un espace de la colonne qui égale en longueur plus du quart de celui 

 qui est formé par les six vertèbres abdominales antérieures. C'est entre les deux dernières 

 vertèbres abdominales, et en arrière de la première caudale, que se fixent les trois osselets 

 interapo])hysaires qui portent les trois rayons épineux de la première dorsale; mais la longue 

 pointe qui est au devant d'eux est produite par le prolongement de l'extrémité postérieure 

 de la carapace, au bout de laquelle est articulé un rayon épineux". It is very probable that 

 A. made his observations on a dried, shrivelled specimen, in which case it is impossible to 

 see the details in question; perhaps the same and only specimen in the Paris Museum, con- 

 cerning which Brühl (5b, p. 51) writes: ". . . . ein kaum ein halb Millim. breiter, vertrockneter, 



häutiger, derber Streifen ist der ganze Rest der Wirbelsäule, an dem Nichts zu erörtern 



möglich". That Günther cannot have seen anything of the spinous processes and interspinous 

 bones in the region in question is quite obvious, otherwise his view of the dorsal cuirass 

 would have been altered at once; he states: "These (six) abdominal vertebræ are extremely 

 slender, the third alone being nearly as long as the whole caudal portion; they have a slight 

 ridge superiorly and inferiorly and on each side; the whole portion lies in the uppermost 

 concavity of the dorsal cuirass ". 



Heckel (15, p. 225) only says that in the fossil A. heinrichi the 6 anterior vertebræ lying 



D. K. D. Vidensk, Selsk. Skr., 7. R:ckke, natuivldensk. of malliem. AW. VI. 2. 13 



