﻿59 97 



el le soiis-opercule forment à eux seuls In partie mobile de l'appareil opereulaire"; and 

 (iüNTHKRs statement {Via, p. 526) thai "the pre- and interopereulum are united into one bone" 

 is just as little correct. 



Concerning the mandible we find in the diagnosis ol' tlie Hcmibrancliii by Smith Wood- 

 ward (36, p. 369): "Mandible simple, each ramus consisting only of two elements (dentary and 

 articulo-angular)'". This is however incorrect; there is an independent angular in all the 

 forms which S. \V. includes under Hemibranchii. 



«; p.66 (28); 



Cope is — so far as I know — the only author who has given any information on the 

 branchial apparatus in Amphisile (6, p. 457i. After first characterising the group Hemibranchii 

 in the following manner (1. c, p. 456): "Superior branchihyals and pharyngeals reduced in 

 number (which as mentioned on p. 42 (4) is incorrect), inferiors separated", he states regarding 

 Amphisile: "Fourth superior branchihyal (i. e. epibranchial IV) and all the superior pharyngeals 

 wanting". Thai all these statements are likewise incorrect appears from the description and 

 figures given by me here. Gii.i. (la, p. 156 and 164) repeats Copes words regarding the bran- 

 chial apjjaratus and Starrs in his diagnosis of the Hemibranchii (30, p. 623) again gives Cope's 

 incorrect statements as follows: "superior pharyngeals and usuallj' elements of branchial 

 arches reduced in number" and p. 625 for the family ''Cenlriscoidea" i. e. Amphisile: "branchial 

 system feebly developed". 



Concerning the gill-rakers in A. punelnlala Kner (21a, p. 534) states: "Die Rechen- 

 zähne des ersten Bogens sind relativ starke nach vor- und einwärts gekrümmte Hakenzähne, 

 die der folgenden Bögen stellen niedere Höckerreihen vor". 



Regarding the hyoid I find the following in Agassiz (lb, p. 276): "Les cornes latérales 

 de l'os hyoïde sont aussi démesurément longues". I am not sure what he means by this; 

 l)erhaps the long ossified tendons which sjiring from the urohyal'.' 



The number of the branch i os legal rays, which are often used by systematists, 

 especially when they can be easily observed, is given as follows: Agassiz (1. c, p. 276) 5, Peters 

 (I.e., p. 335)4; Steindachner (I.e., p. 766) 3 (4?); Günther (I.e., p. 526) 3. Kner (21a, p. 534) states 

 on the other hand that he has not been able to find anj' trace of branchiostegals (in A. piinc- 

 tulala)'. 



5; p. 67 (29): 



Starks cites no earlier account of the pectoral girdle than Günther's comparison of the 

 external characters in Amphisile punelnlala (not A. siriyata, as Starks says) with those in 

 A.scniala, the latter of which is quite ignored by Starks. Nor is there much to be found in 

 the older literature; I know only the following. In Agassiz (lb, p. 276): "Les pectorales.... 

 sont i)ortces par la saillie postérieure du humérus (i. e. : clavicle), auxquels s'attachent les 

 cubitus (i. e. : coracoid) qui se réunissent en avant, comme les apophyses antérieures des deux 

 humérus le font sous la gorge. L'osselet styloïde (i.e.: postclavicle) est derrière l'insertion 

 des pectorales". Günther (14a, p. 526; A.sculata) states: "The humerus (i.e.: clavicle) also 

 contributes to the bony covering of the body ; a long horizontal portion of it extends from 

 the operculum to the base of the pectoral fin; it fits into the shallow notch of the dorsal 

 cuirass mentioned, and is of a lanceolate shape, tapering into a point posteriorly". From 

 the subsequent sentences, which describe the ventral armour (cf citation above, ]).94(56|), and 

 the account of A. punelnlala on p. 528, it is seen that G. has taken the rachis in the 5th ventral 

 plate to be the "coracoid", i. e. the postclavicle, although this can often be seen lying deeper 

 in through the abdominal wall. 



HiLGENDoRF (17, p. 54), wlio rightly denies that the ventral plates have anything to do 



13* 



