i) DR ARTHUR T. MASTERMAN ON THE 



than that of the midwater. At 23 mm. the decreasing midwater curve is met by the 

 littoral, which from here to 32 mm. increases by increments from the midwater, and 

 thereafter the results of distribution and of destruction by enemies cause a slow but 

 sure decrease. The extreme limits of the surface forms are, from the egg to 1 8 mm. ; 

 those of the midwater, from 4 to 39 mm. ; and those of the littoral, from 23 mm. 

 onwards. The intersections of the respective curves are at 8*5 mm. and 22 5 mm. 

 respectively, and these must be regarded as representing the epochs at which the 

 average normally developing cod passes from one division to another. 



It may be noted that the migration downwards from the surface to the midwater 

 is more abrupt and effected more rapidly than that from the midwater to the littoral 

 region, and this is especially so when it is recollected that growth is more rapid during 

 the former period. 



The general facts of this migration have, of course, been well known to scientists 

 for several years, but we are further enabled to say that the average cod lives in the 

 surface water till a length of 8-9 mm. is reached, when it moves down into the mid- 

 water. Here it remains till the proximity of land is reached and it reaches its bottom 

 habitat at about 22 to 23 mm. in length. 



There is no evidence in support of the idea that the young cod reach the bottom 

 at any great distance from land and then migrate inward along the bottom ; in fact, 

 the study of the bottom-net curve in Table II. shows evidence to the contrary. 



The egg and larval stages of the cod have been fully described* and are well known. 

 In the St Andrews laboratory the young larva? have occasionally been kept alive 

 till all the yolk has disappeared. The length of one of these is given by Prof. 

 M'Intosh as 4 mm., and even 4*5 mm. Following upon a description of this stage, 

 the same authority describes briefly a number of transition forms which were caught 

 in the tow-net. In the present paper several of the stages described by him are figured 

 for the first time, and such intermediate stages are added as seem necessary to make 

 a complete series from the close of the larval epoch to the young cod of one inch or 

 more, from which latter stage there is no difficulty in identification. Fig. 1 represents 

 an early post-larval form of 4 '5 mm. (in spirit), probably one of the 5 mm. (fresh) 

 specimens noticed by Prof. M'Intosh. The characteristic larval bands of black 

 pigment have been reduced to two, which are the two post-anal ones. The ventral 

 elements of these two have become fused by the appearance of pigment spots between 

 them. A median ventral line of pigment now extends from the anus nearly to the 

 tip of the tail. This pigment is superficial, and lies at the base of the marginal fin. 

 Dorsally are the upper elements of the two bars and a few scattered pigment spots 

 lying over the brain. There is no trace of the two anterior bars. With the addition 

 of the intense black pigment of the eyes, this is all the pigmentation discernible in 

 .spirit specimens. On the other hand, on clearing, one can readily see a mass of black 



* Trans. Roy. Soc. Edin., vol. xxxv. pi. iii. pp. 812-822. 



