622 NOTES AND MEMORANDA. 



over the projecting part of the scale. The laminated portion is ex- 

 ceedingly thin and slender, and of the future denticles only the chief 

 ones are present, and those in a rudimentary condition (Fig. 3). The 

 central large spine (frequently double) springs directly from the 

 nucleus, and is longer than the scale itself. At the tip it is bent 

 upwards or outwards, a condition which later gives rise to the roughness 

 of the eyed side of the adult topknot. The spines are, as yet, soft 

 and flexible, and when stained with methylen blue display a central 

 stained core surrounded by a clear unstained marginal portion. The 

 smaller spines are bent downwards or inwards at their tips, as if to 

 keep the scale in position. The posterior portion of the scale has a 

 number of black stellate chromatophores scattered over its external 

 surface. 



Phycis blenno'i'des, Brunner. 



In the Annates du Musde cVHistoire Nahirellc de Marseille, tome v., 

 p. 126, Holt has doubtfully ascribed certain pelagic eggs, with single 

 oil-globule, to this species. In a previous number of the same journal, 

 tome iv., Marion figured a larva in all respects similar to that of 

 G. minutus, save that it was smaller, and ascribed it doubtfully to 

 Phycis. Llore recent research has shown that the eggs described by 

 Holt belong to another species of gadoid, and from observations made 

 at Banyuls during the spring of 1898 I found that Marion's conjecture 

 was in all probability correct. The ripe eggs of Phycis, obtained in 

 May, varied from -80 to -88 mm. in size, being slightly smaller there- 

 fore than those of the poor cod.* There was no oil-globule. During 

 the same month the eggs of this species and of G. minutus were taken 

 together in the tow-nets. At first these were all relegated to the 

 latter form, but the constant occurrence of a large number at -80 mm., 

 as well as the observation of the freshly extruded eggs of Phycis, left 

 little doubt that both species were really present. 



During the first two or three days of development of the tow-net 

 eggs, there was no trace of pigment on embryo or zona ; later, a few 

 black spots appeared along the sides of the body and on the head. On 

 hatching, the larva is from 2 to 2-2 mm., and possesses the usual 

 gadoid characteristics. The figure given by Marion {loc. cit., Plate II., 

 Fig. 16) represents it accurately. 



It is curious that Phycis should so closely resemble the poor cod 

 rather than the more nearly related species — the hake, rockling, and 

 ling ; but the herring and its allies have already aftbrded examples of 

 widely different types of eggs in neighbouring species. 



• -QOS mm., Mcintosh and Masterman, British Marine Food-Fishes, 1"0 mm., Raffaele, 

 Mitt. Zool Stat, zu Neapel, 1888. 



