Holt — On the Eggs and Larvae of Teleosteans. 469 



60—70 mill., with an oil-globule 16—18 mm. O. giofredi is regarded by Day 

 [op. cit., vol. i., p. 269) as a synonym of C. julis. Hoffman [op. cit., p. 11) 

 describes the ovum of Julis vulgaris as having a diameter of "75 mm., with an oil- 

 globule of "15 mm., whilst the newly-hatched larva measures l'77mm. 



Coris julis is the only one of these species that is known to be British, and it 

 does not seem to have been recorded from Ireland. Centrolabrus, of whose 

 development nothing is known, appears to be fairly abundant on the west coast. 



I am inclined to regard this form as belonging to some member of the 

 Labridae, and closely allied to Coris. The structure of its notochord presents 

 considerable resemblance to that of Ctenolabrus. Midlus siirmulletus and J/, har- 

 hatus (see Raffaele, op. cit., pp. 20—22) are the only other forms which present 

 any close resemblance to this in their earlier stages, but, besides differences of 

 dimensions, there is in their ova a layer of cortical yolk segments, and the anus in 

 the larva is nearer to the yolk. 



Species VI. 

 (PI. XLVIII., fig. 17.) 



This species is represented by only a single example, which was taken in the 

 surface-net in Inver Bay on the 25th June. 



The diameter is l'13mm., and there is a single oil-globule {o.g.) of •21mm. 

 The yolk is clear and homogeneous, and there is no colouration of any sort. 

 Whilst agreeing with another unidentified egg, species VII., in measui'ements, it 

 is at once distinguished from it by the large size of the perivitelline space {p. s.), 

 and by the entire absence of pigment at a somewhat advanced stage. 



My observations deal only with the stage shown in fig. 17. The embryo is 

 about l'14mm. long, of which "30 belongs to the precaudal region. The eyes are 

 large. I could not make out the otocysts, or the heart. The oil-globule is at the 

 posterior end of the yolk (y.), which is '66 mm. long and "93 mm. broad. 



In the presence of a single oil-globule and a large perivitelline space this egg 

 approaches that of the pilchard (see RafPaele, op. cit., p. 55, and Cunningham's 

 later Paper, p. 43), but differs from it in dimensions and in the absence of the 

 reticulation of the periblast, which is characteristic of clupeoid ova. 



Species VII. 



(PI. XLVIII., fig. 18 ; PI. XLix., figs. 25, 25a.) 



A few ova, which appeai'ed to belong to the same species, at the same stage of 

 development, occurred in the surface-nets in Inver Bay on the 25th June. 



3U2 



