490 LOUIS AGASSIZ. 



medusae has already been recognized. I do 

 not here allude to their primitive origin, but 

 simply to the general fact that among radi- 

 ates the embryos of the higher classes repre- 

 sent, in miniature, types of the lower classes, 

 as, for instance, those of the echinoderms re- 

 semble the medusae, those of the medusae the 

 polyps. Having passed the greater part of 

 last winter in Florida, where I was especially 

 occupied in studying the coral reefs, I had the 

 best opportunity in the world for prosecuting 

 my embryological researches upon the stony 

 corals. I detected relations among them which 

 now enable me to determine the classification 

 of these animals according to their mode of 

 development with greater completeness than 

 ever before, and even to assign a superior or 

 inferior rank to their diJfferent types, agreeing 

 with their geological succession, as I have 

 already done for the fishes. I am on the 

 road to the same results for the moUusks and 

 the articulates, and can even now say in gen- 

 eral terms, that the most ancient representa- 

 tives of all the families belonging to these 

 great groups, strikingly recall the first phases 

 in the embryonic development of their suc- 

 cessors in more recent formations, and even 

 that the embryos of comparatively recent 



