468 Holt — On the Eggs and Larvae of Teleosteans. 



is separated by a considerable interval. The imperforate anus(«.) is marginal. 

 There is a small urocyst {u.) in the usual position. 



The eyes are of moderate size, unpigmented. The pineal (;;«.) is visible ; the 

 cerebellar fold is small. The otocysts (o^.) are small, and somewhat conical 

 dorsally, and as yet remote from the eye. The pectorals have not appeared. 

 The dorsal marginal fin commences behind the otocysts, and reaches its greatest 

 height, about -18 mm., a little behind the anal region. The ventral marginal, 

 slightly less than the dorsal, is indented towards the anus, as is the pre-anal 

 segment (jo.rt./.). The caudal is broad and rounded, and slightly spatulate at 

 this stage. 



The pigment is all black, and is confined to the head and trunk. Largish 

 stellate chromatophores are distributed sparingly and somewhat irregularly over 

 the anterior two-thirds of the dorsal surface. A small chromatophore marks the 

 hind-wall of the urocyst. A little behind the last dorsal, there are two stellate 

 chromatophores on the ventral edge of the body, and a few small stellate chroma- 

 tophores occur ventrally and laterally in the extreme caudal region. 



A day later the larva (fig. 44) is only 2-38 mm. in total length. This decrease 

 is due to the partial absorption of the yolk and withdrawal of its anterior 

 extremity, with the oil- globule, to a point in rear of the fore-brain. The same 

 cause has operated to increase the downward flexure of the brain (a condition the 

 reverse of that which is met with in the development of most teleostean larvse), 

 thus further decreasing the pre-anal length, which is now only 1-01 mm. 



The post-anal length has, however, gained -06 mm. The yolk is now bluntly 

 ovoidal. The gut shows a slight dilatation in the middle of its length (5.), and its 

 tubular region extends a little further into the rectum. The pectorals (;^./.) have 

 appeared in the usual position ; they are as yet very small, little more than 

 prominences of the lateral epiblast. A great deal of the pigment noticed on the 

 previous day has disappeared; the head is altogether destitute of it, and the dorsal 

 chromatophores are fewer in number, whilst some of them appear to have migrated 

 to the lateral region. The extreme posterior end of tlie caudal region is slightly 

 upturned, and the subnotochordal pigment is now dendritic, and extends on to the 

 marginal fin. Embryonic caudal fin rays have appeared. The heart at this, as 

 at the previous stage, occupies the usual position, and beats actively. The 

 structure of the notochord (fig. 45) is somewhat characteristic. The cells 

 (vacuoles) are large, and sometimes one, sometimes two, or even three occur in the 

 same plane of transverse section. I was not successful in rearing this species to 

 an older stage. 



The ova and larvae of this form present a remarkable resemblance to those of 

 Coris j'ulis, Coris giofredi, and Julis turcica, as described and figured by Raffaele 

 (ojo. cit., p. 35). Raffaele's ova, however, are smaller, having a diameter of 



