178 Mil. J. T. CUXXINGHAM OX [Feb. 19, 



partial metastvophe does not apply. In this case we must conclude 

 that some other explanation is to be sought, or we may suppose 

 that the boundaries between the determinant groups in the ovum 

 are not deiinite, and that the pigment determinants displaced to 

 the right side have extended to the head-region. 



It may be objected that the persistence of colour on the lower 

 side of an ambicolorate Turbot is inconsistent with my views of 

 the action of light, that if pigment were produced on the upper 

 side it ought to disappear from the side turned to the ground. 

 This objection is of little weight, for my experiments show that it 

 is easier by means of light to produce some pigment where it was 

 previously absent, than to abolish it when it is present, by cutting 

 otf the light. This is what might be expected, for in the evolution 

 of a flat-fish pigment has only recently disappeared from the lower 

 side, in consequence, as I believe, of the absence of light ; and 

 therefore the pigmentless condition is not very strongly inherited, 

 and pigment is produced after a comparatively short exposure to 

 li,o-ht. The positive character on the other hand, the presence 

 of pigment, has existed not only since the flat-fish was evolved, 

 but in a long line of ancestors before that, and therefore it would 

 probably take several generations to cause the pigment to disappear 

 completely by cutting ofF the light. It is quite possible that when 

 the lower side is congenitally pigmented, some proportion of the 

 pigment is lost in consequence of the absence of light, but such a 

 loss wovild not be obvious to observation and would be difficult to 

 demonstrate. Obviously a small amount of pigment appearing 

 where there was none before is evident at once, but the disappear- 

 ance of a small proportion from a strongly pigmented surface 

 makes no apparent difference to the colour, and there is no means 

 of measuring the amount of pigment for comparison in different 

 cases. There can be no doubt concerning the presence of a single 

 sheep in a field, but it is much more difficult to decide whether 

 there are a thousand or 999 in a flock. 



It has long been known that in Pleuronectidie generally, and 

 especially in Rhombus maximvs, there is a marked correlation 

 between ambicoloriition and the malformation of the dorsal fin 

 which occurs in the specimen described in this paper. It seems to 

 be generally supposed that in such specimens the dislocated eye 

 has not completed its change of position, and being on the edge 

 of the head instead of on the upper side, prevents the usual growth 

 forwards of the base of the dorsal fin. The condition is regarded 

 then as, like the ambicoloration, a reversion on the part of the 

 eyes and skull towards the primitive symmetry. Although I have 

 not fully investigated the structure anatomically, it is my opinion, 

 from external observation, that the eyes and skull are normal and 

 that tlie peculiarity is merely due to a want of that attachment 

 which normally occurs between the base of the fin and the skull, 

 alon^- the united ectethmoid or prefrontal, and frontal bones. The 

 view I have suggested seems to me to give a better explanation 

 of this abnormality than has hitherto been proposed. If the head 



