LOCALIZED STAGES IN DEVELOPMENT. 141 



bjsexiiiil reproduction. The question arises, could such new forms originate from such 

 variations as are seen in locahzed stages in development of either a progressive or regres- 

 sive character. It is probable that variations may at least occasionally originate in this 

 manner in the state of natui-e. Professor Bailey ('96) states that seedlings of varieties 

 which have originated from bud-\ ariation are quite as likely to reproduce the variation as 

 those raised from varieties originated by seed-variation. It is (piite possible, therefore, 

 that indiviihials wliich diifer from the type originated from seed borne on localized 

 branches which dilfered in the same direction. 



In plants, variations in localized areas have been noted in many cases, as in the Tulip- 

 tree, Sassafras, Platanus, Pitch Pine, Red Cedar, etc. In animals, areas of localized varia- 

 tions have been noted in Palaeozoic Echini. Such variatitjns in localized parts as observed 

 are not haphazard and heterogeneous in kind. They are all perfectly definite, being either 

 atavic or progressive variations, and can be correlated with similar characters seen in the 

 young or in fossil representatives, or else in more advanced species of the group to which 

 the animal or plant belongs. Occasional variations cannot be correlated with the ontogeny 

 or phylogeny of the type, such as the split leaves in Libocedrus, Fig. '.>1. These may 

 properly be considered as monstrosities, comparable to fan-tail pigeons, top-knot fowls, 

 double flowers, cut-leaf birches, etc. 



In aninuils which reproduce asexually by budding, as Hydrozoa and Actinozoa, it 

 seems that the bud may be considered a locaUzed stage (Galaxea, PI. 24, figs. l()(j-l()'.)) . 

 The bud has not the stages seen in early embryonic development from the egg, but re- 

 peats in general the later stages seen in such an ontogeny. A bud is not a new individual 

 in the full sense of the word, but is an outgrowth from an older individual by a special 

 localized development. 



In animals which, during growth, acquire an addition of similar parts, there is often 

 a strongly marked ontogenesis of such parts, which is parallel to the condition seen in the 

 young or in simpler, less specialized, and ancestral types. Echinoderms, which during 

 growth acquire additional plates, in certain cases show evidence of such localized ontogen- 

 esis. Such are the addition of new plates to the stem in the recent crinoids, Metacrinus, 

 Fig. 124, and Pentacrinus, the young newly added plates being comparable to the plates 

 of the whole stem in Jurassic representatives, Fig. 125. 



In the Echini (Strongylocentrotus, Figs. 112-116, and Arbacia, Figs. 110-111), new 

 plates are added at the dorsal" border of the corona. Such young plates repeat the char- 

 acter seen in all the plates of a very young individual, and are similar to the plates of 

 Cidaris, which is a primitive type in this group. As there is an ontogenesis of the indi- 

 vidual plate, so a series of successively older plates presents successive phases of ontogenies ; 

 as a result, in passing from the dorsal Ijorder of the corona ventrally, a series of stages 



