1906.] THE SKULL OF A YOUNG RIBBON- FISH. 547 



Otherwise one might state that these denominators decrease as 

 the fish increases in length. 



The really important differences in these relations occur in 

 those cases in which the greatest height is a factor — which is veiy 

 much less in proportion to the length in the young than in the 

 adult. 



If, then, " R. parheri " be the young of R. glesne, we have not 

 only a gi'eat change taking place during growth, in the pi'oportions 

 of all the parts, especially the height-length I'elation, but a 

 diminution in the total number of fin-rays in the dorsal fin, and 

 the breaking up of tiansverse coloured bands into oval and 

 circular spots, and also the appearance in the fore part of the body 

 of much darker iriegular streaks. 



"We know that somewhat similar changes do take place in fishes 

 belonging to the family in which Regalecus is included. In 

 Trachypterus, for example, such changes are illustrated in 

 Giinther's 'Study of Fishes,' p. 521, and apparently Liitken 

 had already expressed the opinion that such growth-changes 

 would occur in Regalecxis (3. p. 294). 



I was disappointed in finding so little about this subject in the 

 recent volume on Fishes in the ' Cambridge Natuiul History ' ; 

 but in the systematic portion Boulenger (p. 714) remarks : " The 

 life-histories (of Tajniosomi) are still very imperfectly kno'^vn, and 

 gi-eat changes of form take place during growth ; " but nothing 

 further is said about the matter. 



I have been unable, owing to the poverty of our libraries in 

 New Zealand, to ascei'tain whether any, and if so what, work has 

 been done on changes in the detailed structure of the skull diu-ing 

 growth. The references to be found in Wiedersheim, or in 

 Ziegler's ' Vergleich. Entwick. d. nieder. Wii-belthiere,' refer 

 only to embryonic changes, so far as I have been able to ascertain 

 fi'om abstracts in the ' Zoolog. Jahresbericht.' 



The present contribution, together with my previous ai'ticle on 

 the external form of R. parheri^ is a step in this direction, if this 

 fish be, in fact, the young of R. glesne. 



II. Description of the Skull. 

 (By W. J. Dunbar.) 



A. The Bones of the Upper mid Lower Jaw. 



The Premaxilla consists of two regions, namely, a thin plate 

 lying at the side of the oral aperture (pnix.), the " alveolar portion" 

 of Prof. Parker, and a long dorsally situated " nasal process " 

 (p7nx.'), which is connected with its fellow of the other side by a 

 laterally compressed plate of cai'tilage (pmx.", PI. XXXVIII. 

 fig. 1 and PL XXXIX. fig. 3). This process extends back over the 

 cranium on the dorsal surface, lying in a cartilage-lined "anterior 

 dorsal " groove in which it can slide to and fro. In R. glesne the 

 bone has relatively a much shorter- nasal process, the two regions 



