﻿94 56 



out a series of misinter])retations. Apaii IVoni lliese, which I sliall point out directly, and 

 from various otlier, less correct data which will be dealt with later, Günther has tiescrihed 

 this form in a more detailed manner than most of the other authors, not exceptinj^ Agassiz, 

 and on many points Günthers account is good and correct so far as it goes. The dorsal 

 cuirass is described (p. 526) correctly in all essentials; that G. counts 6 i)lates in the lower 

 row (the lateral) has arisen from the fact that he has not recognised the supraclavicle as 

 such but takes it to be a plate of the cuirass. On the other hand, he has rightly recognised 

 the part of the clavicle as such id in my fig. 1, PI. I) and keeps it distinct from the true dorsal 

 armour. It may be specially mentioned that G. was the first who suspected the true nature 

 of the dorsal spine, as he say.s: "The long moveable (should be immoveable!) spine in which 

 the cuirass terminates is evidently an interneural, since, in the second species, the first sjjine 

 of the dorsal fin is joined to it". It is only the dorsal cuirass that G. regards as true armour; 

 this is shown already in his diagnosis of the genus (]). 524) where it is only stated: "Body — 

 provided with a dorsal cuirass which is formed by portions of the skeleton'; and later under 

 the description of A.sculala (i). 525) it is said: "The whole head and back are cuirassed with 

 smooth bony plates, whilst the abdomen is covered with a very tough skin ", and in more 

 detail on p. 526: "The inferior half of the side of the trunk is covered by a transparent tough 

 covering, which passes into a broad cutting fringe interiorly; this fringe extends the whole 

 length of the abdomen, and the whole covering is supported by the coracoid and by seven 

 or eight ribs'". G. has thus in the first place not noticed the boundaries of the plates from 

 one another and in the second quite misunderstood their structure; his coracoid and the 7 

 or 8 ribs are in fact merely the rachides on the lateral plates (the "coracoid" is the rachis 

 on the 5th plate; see later). This misinterpretation is in comi)lete agreement with the whole 

 of Gönther's incorrect views on the mor])hology of the dorsal armour, expressed (1. c, p. 527*) 

 in the following words: "I am of o|)inion that the dorsal cuirass is not a dermal production, 

 but formed by modified |)arts of the cmloskeleton; its composition, the number and ])osition 

 of its single parts, and, finally, the first dorsal s|)ine, which in A. piinclalata is so singularly 

 attached to it, favour this opinion. The jilates which occujjv tlie vertebral line would corre- 

 spond to the neural spines, and the lateral plates on which the ribs are sus])ended, to the 

 parapophyses. Aniphisile may be considered as a Chelonian form among fishes". It appears 

 quite extraordinary, that the ])osition of the musculature in under the sujiposed spinous jiro- 

 cesses, covered by these (which may be seen indeed without preparation) and a "coracoid" 

 (i. e. postclavicle) in front of the pectorals etc., did not raise some doubt in Gönther's mind 

 as to the correctness of such an interpretation; it is obviously the Chelonian idea which has 

 controlled and confused the account. 



In a notice by Steindachner (31) of about the same time we find the following remark: 

 "Bleeker und andere Ichthyologen unterscheiden in der Beschreibung von Amphisile nur 

 Bauch- und Hückenschilde, während doch Bauch-, Seiten- und Bückenschilder vorhanden 

 sind, welche durch wahre Naht mit einander verbunden sind". No further explanation is 

 given, but S.'s "Seitenschilder" must be the lower row of plates in the dorsal cuirass, just as 

 with Hilgendorf. LOtken (24a) speaks of sutures between the ventral plates of this and that 

 number, but at the same time of the number of "ribs" occurring in the ditferent sjjecies, 

 from which we may suppose that he agrees with Gönther's views. The only protest against 

 this that I have found anywhere occurs in Hilgendorf (17, |). 54) in the following words: "Die 

 Bauchplatten und deren knöcherne Verstärkungslisten können nicht, wie dies wolil geschehen, 

 mit Rippen in Verbindung gebracht werden, dazu ist ihre Zahl schon zu gross, während die 

 knöchernen Seitenplatten in der Zahl mit den Wirbeln corresjjondiren und auch, mit Aus- 

 nahme der letzten, mit ihnen in continuirlicher Verbindung stellen". This jirotest seems to 



* Tlie same is repeated in tlie "Introduction" (14 b) practically in tlie same words. 



