COTTUS GKCENLANDICITS. 7 



have appeared in successive years. In fact, whenever we were particularly 

 anxious to find a given stage, the date of occurrence of former years formed 

 an infallible guide in our search. This regularity in the appearance of 

 pelagic animals is not limited to young fishes, but we find that the time 

 of spawning of t he majority of marine animals and their rate of develop- 

 ment arc practically identical year after year. We may quote as instances 

 the appearance of Plagusia, Agalma, Arachnaetis, Balanoglossus, and hosts of 

 other embryos, on fixed days, as examples of the close interdependence we 

 find between the season and the temperature of the sea with the growth of 

 pelagic life. 



Cottus grcenlandicus, C. & V. 



Cf. Agassiz's Young Fishes, Part III., PI. III. ami PI. II. Figs. 1 and 2. 

 Plate I. Fljis. 1-5. 



The eggs referred to this species are found throughout the summer 

 months, but are most abundant in July. They are easily distinguished with 

 the naked eye by their large size (1 to 1.1 mm.), and especially by the pres- 

 ence of scattered oil-globules, the number of which in a single egg may 

 vary between ten and forty. These oil-globules are sometimes distributed 

 more or less evenly in the periphery of the yolk (Figs. 1 and 3). and some- 

 times a tendency to cluster is shown by some or all of them (Fig. 2). In 

 size they vary from .02 mm. to .10 mm. No instance of coalescence into a 

 single globule has been noticed. Their peripheral position shows that their 

 place in the yolk is not determined by specific gravity alone. 



Yellow pigment begins to appear in this species three or four hours 

 before the closing of the blastopore, about the time the first somite becomes 

 plainly marked off, and almost simultaneously with the minute cluster of 

 vesicles which are destined to coalesce, and thus give rise to Kupffer's vesicle. 

 The pigment is at first extremely .pale, and confined to a few mesoblastic 

 cells along each side of the embryo. In the course of an hour the number 

 of yellow pigment-cells is much increased, and a lew black pigment-dots 

 make their appearance. By this time some of the pigment-cells of both 

 colors have wandered away from the lateral mesoblastic masses of the 

 embryo, and appear as isolated amoeboid cells between (be ectoderm and 

 the layer which we have called the periblast. 



Thus far not a single pigment-cell lias appeared anywhere in front of 



