DEVELOPMENT OF TBACHINUS YIPEEA. 275 



then tliat the egg rises to the surface. Although the number 

 and size of the oil-globules is variable, this is only witbin 

 certain limits ; and I have never found an egg of Trachinus witb 

 only one large oil-globule, as appears to be the rule with the 

 majority of other species of pelagic eggs hitherto described. 



The eggs ai-e laid in the night ; but at what time I am 

 not certain. We have watched the fish up to 1 a.m., and 

 resumed watch again as early as 5 a.m., but have never been 

 able to catch them in the act of ovipositing. They are pro- 

 bably laid in the very early hours of the morning, just before or 

 after daybreak, as we always found them well advanced in the 

 segmentation stage by even 5 a.m. I have this year, however, 

 had an opportunity of studying the segmentation process from 

 its commencement, in a few eggs laid in the beginning of April 

 and at a temperature 9 or 10 degrees lower than at their normal 

 time of appearing last year, and consequently considerably re- 

 tarded in development. The eggs, being at the surface of the 

 water, are naturally more affected by the temperature than would 

 be the case otherwise; so that in order to estimate fairly the rate 

 of development the temperature of both room and water must be 

 taken into account. The temperature of the water in my aqija-? 

 rium varied during the months of June and July from 54 to 58 

 and 60, and that of the room from 54 to 62, and during the last 

 two days it went up to 65. 



A comparison of the various times at which the embryo deve- 

 loped certain organs or structures made in vq-rioug batches of 

 eggs at various temperatures, showed that a difference of about 

 2 degrees would retard or accelerate from 9 to 12 hours in the 

 early stages, and a whole day for hatching. I propose, however, to 

 leave the consideration of this part of my subject to a future paper. 



Egg-Membranes. — In the fertilized floating ovum the follow- 

 ing investing membranes can be distinguished : — 



(1) An exceedingly thin membrane showing only as a fine line 

 under the 14nch objective, hyaline and apparently structureless 

 and non-perforate. This is the vitelline membra7ie according to 

 the definition of Balfour. 



(2) Within this, and occasionally separated from it by a space, 

 is a much thicker membrane, the zona radiata ; but I have maae 

 no observations on its structure. This is separated from the 

 yolk by a space, the " breathing-chamber" of Eansom, which is 

 only small in the species under discussion. Eyder describes only 



