THE SAND-EEL FAMILY. 307 



Those procured in the beginning of January and measuring 

 between 6 and 7 inches were also fully ripe, the abdomen being 

 distended in both males and females. In the dead examples 

 the eggs measured -8 mm. and the oil-globule 2 mm. More- 

 over the colour of the egg was pale greyish and that of the oil- 

 globule, in certain light, pale greenish. Whether this variation 

 in colour was due to post-mortem changes was not clear. 



In St Andrews Bay there are found every year at the 

 beginning of March immense numbers of small larvse, which 

 occur in the bottom tow-nets. They have been figured and 

 described by Mcintosh and Prince as larva D, and later they 

 have been definitely identified by the former as young sand- 

 eels, and the slightly yellowish colour in the oil-globule leads 

 us to infer that they are of the 'lesser' species. Fig. 6, 

 Plate XII, gives a lateral view of one of these little larvae. 

 They are about ^th inch (5 to 6 mm.) in length. The yolk 

 is small — being nearly absorbed, but the pale oil-globule is still 

 present, contained in a thick layer of protoplasm. The larva is 

 elongated (Plate XII, fig. 5) with a vent at about fth of length 

 from head to tail, and has a little dark pigment along the 

 ventral edge. In some points it resembles the young gunnel 

 but may be distinguished by the arrangement of the black 

 pigment. At a little later stage than in fig. 6, a double ventral 

 row of black spots is found along the abdomen, single behind 

 the vent and in the region of the pectoral fins. The marginal 

 fin is continuous, while the mandible protrudes in front of the 

 head. There are small transparent pectoral fins and a little 

 black pigment, consisting of a row of black spots below the 

 abdomen and another row dorsal to the intestine. 



Such then is the little larval sand-eel which occurs in 

 immense quantities upon the sandy bottom, in the company of 

 numbers of larval herrings, which are more elongated, besides 

 differing in other points, and of young arrow-worms, animals 

 which are also of an elongated outline. When we recollect the 

 shape of the adult sand-loving sand-eel and of the lancelet we 

 are inclined to believe that the elongated form of the young 

 sand-eels and herrings is also probably a special adaptation. 

 The suddenness of the appearance of these larvae and their 



20—2 



