6 PROF. G. B. HOWES ON THE SKELETON AND [Jan. 18, 



what I imagine may represent the cartilage in question, and that, as is 

 here the case, in irregular fins. Be it, as it there exists, what it may, 

 its characters in the fin figured by me are still further noteworthy. 

 The entire fin-skeleton (fig. 1), with the exception of this bar and the 

 proximal mesomere, is very slender and leaf-like ; the two elements 

 just named (which, be it remembered, are in direct connexion) are 

 relatively massive and much thicker and more powerful than the 

 rest. The bar r, instead of being ellipsoidal in transverse section, 

 as is invariably the case with even the most powerful parameres, 

 is expanded along its free border in a manner strikingly suggestive 

 of the metapterygium as it exists in many Elasmobranchs. It is 

 segmented into a main piece and two small terminal ones, and 

 appears, at first sight, to represent an element of greater importance 

 than an ordinary ray. 



The fact that this new element appears in "irregular" fins, taken 

 in conjunction with the fact that no such structure has hitherto been 

 recorded for a " regular " fin, appears at first sight to detract from its 

 novelty. Before proceeding further, therefore, three questions must 

 be met : — 1. How far is the fin under discussion abnormal? 2. Can 

 the existence of the new element be demonstrated for a more normal 

 fin ?, and 3. If so, under what structural conditions does it exist ? 

 Giinther, in his original description of the Ceratodus fin, described 

 (14, p. 532) certain " slight irregularities " in the distribution of the 

 rays. Huxley (19, p. 47), commenting upon these, remarks that 

 they are "in respect of the median pieces .... constant peculiarities 

 of no small importance." Davidofi'i'Z, p. 126) describes the stem of 

 the pelvic fin as consisting of a row of pieces " deren Zahl bei den 

 verschiedenen Individuen betrachtlich variirt ; " he adds — " nirgends 

 fand ich ein so unregelmassiges Verhaltniss derselben zu einander, 

 wie es Giinther auf seiner Figur abbildet." Other writers have 

 observed this irregularity, and the last of them (Schneider) has 

 formulated the distribution of the parameres of both fins. He states 

 (23, pp. 521-22), "bei der. Brustflosse sitzt dorsalwiirts am zweiten 

 IdIs elften Gliede des Hanptstrahls, und zwar an der distalen Gelenk- 

 flache, je ein Seitenstrahl. Ventralwarts sitzen am zweiten Gliede 

 des Hauptstrahls hinter einander fiinfSeitenstrahlen, am dritten und 

 vierten Gliede je zwei, an den folgenden einer. Bei der Bauchflosse 

 tragen dieGlieder des Hauptstrahls ventralwarts je einen Seitenstralil, 

 dorsalwarts je zwei Seitenstrahlen." I have taken some pains to 

 test the reliability of this very definite statement, and am in a 

 position to assert with equal assurance that the only constant 

 character as yet recognized is the attachment of one ray to tbe pre- 

 axial border of each pectoral mesomere {cf. figs. 5 & 6). Even in so 

 modified a fin as that of fig. 5, where several of the parameres are 

 branched and two are directly confluent, this rule holds ; and in no 

 regular pectoral fin yet examined has an exception to it been found. 

 I give below a table of average distribution of the parameres of those 

 segments dealt with by Schneider, calculated out from observations 

 made upon eight pectoral and ten pelvic fins. 



