124 CEESSWELL SHEAEER, WALTER DE MORGAN, H. M. FUCHS. 



later and less variable stages. The chief index of paternal or maternal 

 influence has been the skeleton ; but this, unfortunately, exhibits a 

 large amount of irregular variation dependent on small metabolic 

 changes, a variation in many cases quite independent of heredity. 



The present work was commenced, therefore, with the object of 

 discovering fixed specific characters, into which we felt certain no 

 irregular variation entered. After considerable investigation we came to 

 the conclusion that these could only be looked for in the later period of 

 larval life. For in our experiments we have found that every culture 

 jar had its own rate of development and showed minor variations 

 with regard to the early larval characters, such as skeleton, pigmenta- 

 tion, and shape, so that it was plain that no definite results could be 

 hoped for from the investigation of young stages alone. 



Thanks to the methods elaborated by Dr. Allen (1) of rearing marine 

 larvse and of feeding them on pure cultures of diatoms from which 

 bacteria and infusoria are as far as possible eliminated, and of keeping 

 the larvae in sterilized sea-water, it is now possible to rear the Plutei 

 of Echinoids with great facility through metamorphosis to the young 

 fully formed Sea-Urchin. * This suggested to us the idea of making a 

 full investigation of the later stages of the pure and hybrid forms, 

 and we hope to show that we have found in them immutable and 

 distinct specific characters which give much more definite evidence 

 than those hitherto used in this work. 



We have chosen Echinus acutus, E. esculentus, and E. miliar is as the 

 three commonest forms on our coasts, and also for the reason that we 

 already possess in MacBride's work (1) a clear account of the main 

 features of the normal development of these species. Of these three 

 forms, E. acutus and E. esculentus at Plymouth are found in deeper 

 water, while E. miliaris is a shore species. It is therefore highly 

 probable that laboratory conditions are more favourable to the latter 

 than to the two former ; and this would seem to be borne out by the 

 fact that E. miliaris develops, under laboratory conditions, much more 

 quickly than the other two, and that in all crosses into which E. miliaris 

 enters the rate of development is accelerated. 



While there are marked specific differences between E. miliaris-f 

 (which has by some authors been placed in a separate genus, Far- 



* While previous investigatorshave unsuccessfully attempted to rear hybrid Plutei through 

 metamorphosis, Doncaster (5), making use of the cross Strongijloccntrottis lividus S x 

 Echinus viicrotuherculatus ? , seems to have been the first to have accomplished this. 

 One of the present authors (Shearer) had the opportunity in 1902 of examining these 

 hybrids at Naples. 



t It may be here noted that E. miliaris, on the Devonshire coast, exhibits several 

 distinct varieties, which differ chiefly in size, in length of spines, and in pigmentation. 



