THE EXPERIMENTAL HYBRIDIZATION OF ECHINOIDS. 139 



miliaris $ which had been fertilized in water treated respectively with 

 alkali and acid. It will be seen that they differ in no essentials from 

 the hybrids raised in normal sea-water. The inheritance of the 

 maternal pigment masses, absence of posterior ciliated epaulettes and 

 absence of posterior pedicellaria is unchanged. 



VII. THE CHAEACTERS OF THE YOUNG SEA-URCHIN. 



In all cases the young fully formed Urchins, in superficial features, 

 such as the character of the spines, pigmentation, etc., show a motherly 

 influence, but so far we have not made a close examination of the 

 number and structure of the test plates. The young Urchins always 

 secrete a thick layer of mucus on the surface of the test, which 

 effectually prevents any examination of the plates in the living state, 

 and any such examination necessitates killing the hybrids. We wish 

 to reserve for a future occasion our statements on this head. 



There is this important feature about the tube feet of E. miliaris, 

 to which attention has been drawn by MacBride (11). He has shown 

 that the young E. esculentus and E. acutus walk by means of five tube 

 feet, each forming the termination of one of the radial canals, and 

 each provided with a sucker, in the centre of the disk of which is a 

 sense organ consisting of elongated cells. In each radius there are in 

 addition the rudiments of a pair of tube feet, which are mere buds in 

 these species, but in E. miliaris these accessory tube feet are functional 

 at metamorphosis. So we have this important distinction between 

 the normal just metamorphosed E. miliaris and E. csculentus or 

 E. acutus, that the former has five principal tube feet and ten 

 secondary ones, while the two latter have only the five principal ones, 

 the secondary ones being entirely rudimentary at this stage. This, 

 therefore, offers a definite index of parental influence after metamor- 

 phosis. On this head also we wish to defer our statements, to in- 

 corporate them later with the evidence offered by the study of the 

 structure and number of the hybrid test plates. 



YIII. SUMMARY OF CONCLUSIONS. 



1. As the result of extensive investigation of the early larval 

 history of our various crosses, we have come to the conclusion that 

 these are too variable to afford any definite evidence of parental 

 influence, and especially is this true with regard to the skeleton, 

 lieretofore considered the chief index of inheritance. 



2. What little evidence we have been able to derive from the study 

 of a very large number of skeletons of both normal and hybrid crosses, 

 seems to show that, while the paramount influence is always maternal, 



