8 Dr. C. F. Llitken on the Changes of Form in Fishes 



in a whole series of well-marked cliaracters, but that the 

 generic separation between the Lepturi [Trichiurus lepturus 

 and the allied species) and the Eupleui^ogrammi [T. muticxis) 

 proposed by Gill is unnecessary, and therefore to be rejected. 

 The specific difference between T. lepturus of the Atlantic 

 and T. haumela of the Indian Ocean seems to me not to 

 repose upon a very solid basis ; and how far the latter is 

 really a species distinct from T. savala is a question which I 

 shall also leave undecided. But there are two points which 

 merit attention, namely : — 1, that T. mutictis also occurs in 

 the Atlantic, where it had not hitlierto been indicated (I have 

 before me a specimen from Cuba which I am unable to dis- 

 tinguish from those from Tranquebar) ; and 2, that in a 

 Trichiurus (hawnela?) 52 millims. long, from Java, I have 

 found, in the place of the ventrals, which are usually deficient, 

 two denticulated spines 2 millims, in length (just as, for 

 example, in the young Prometheus atlanticus) . These spines, 

 which evidently represent the ventrals, probably exist in all 

 the young Trichiuri'^ but they persist only in T. muticus, in 

 the shape of small rudiments in the form of scales, and disap- 

 pear entirely in the other species. 



The Gempylides and the Thyrsitides present an osteological 

 peculiarity which has generally passed unnoticed ; I refer to 

 a system of dermal ribs, or subcutaneous accessory ribs, com- 

 posed of delicate bony filaments, placed pretty close together, 

 which are directed backwards, and both upwards and down- 

 wards, and start from the median line on both sides of the 

 body, forming acute angles with each other. I have observed 

 them in Thyrsites atun {chilensis}, in Nealotus tripes, and in 

 both large and small examples (down to a total length of 

 78 millims.) of Ge77ip7/Ius serpens. I have been able to study 

 the last-named species, which is essentially pelagic and rare 

 in museums, in all the pliases of its development, from a 

 length of about 1 metre down to a total length of only 



9 millims,, a phase which must have almost immediately 

 succeeded the hatching of the egg. The species established, 

 namely G. coluber (genus Zyphothyca of Swainson) and G. 

 or Pj-ometheus [Nealotus) ophidianiis.^ Poey, do not differ 

 specifically from G. serpens ; but the characters of this genus 

 are not correctly indicated, even in the most recent works. 

 Thus the skin is not entirely destitute of scales : we find in 

 it a rudiment of the " corselet," not only behind the eyes, but 

 also at the root of the tail ; and from this point the clothing 

 of scales continues in part for a small extent along the infe- 

 rior lateral line, in part for a greater extent along the back, 

 forming a band which is limited inferiorly by the superior 



