202 PROCEEDINGS OF THE ACADExMY OF [April, 



bud -or gonophore be sufficiently specialized in structure from the 

 polyp condition to be regarded as another individual and not merely 

 a part or organ of the polyp. For, granting the correctness of the 

 general view that such gonophores represent modified medusae, 

 which at an earlier racial period led an independent, free-swimming 

 existence, they may in extreme cases lose all the original medusoid 

 characters except that of producing eggs, show no independence of 

 the polyp stalk, and, in fact, be more correctly considered organs 

 (gonads) of the latter than separate individuals. This would be a 

 case where a part once cut off as a sej^arate individual with an 

 independent existence has been gradually reduced to the state of an 

 organ. In such a case, then, where the gonophore shows no inde- 

 pendent existence, but is a part of the polyp stalk, the whole 

 organization — i. e., the polyp stalk with its gonophores — should be 

 the one classified, because all together represent one organization. 

 This whole organization should be classified when its reproduc- 

 tive organs, the gonophores, are in functional activity, that being the 

 period of the greatest perfection of the organism. The difficulty in 

 putting this principle to practice is to be found in those cases where 

 it is hard to determine Avhether the egg-producer is to be considered 

 an independent medusa or an organ of the polyp stalk. But at 

 present we are considering only the principles of classification ; two 

 j)rocesses or states which in their extreme conditions are easily 

 separated, are often found in natural phenomena to be connected by 

 gradual steps ; this is to be expected in any gradual evolution. 

 Each case difficult of decision must then be considered in turn. 



The principle, then, in cases of pronounced change of genera- 

 tion, whex-e the successive individuals are structurally different, is 

 to classify the organism at the end stage of each cycle. A cycle is 

 in a sense a circle, without beginning or end ; yet a little thought 

 convinces that there must have been a commencement to every cycle. 

 In the case of a Metazoan, where from the egg develops the mature 

 organism, and from that an egg again, we commonly speak of the 

 mature organism as the end and the egg as the beginning ; and with 

 right, since development is from the more generalized to the more 

 specialized. Or regarding only the germ-cells of a Metazoan, 

 apart from the soma, we find also a cycle : n generations of ovo- 

 gonia, a stage of synapsis of the chromosomes, a stage of matura- 

 tion, a stage of conjugation (fertilization), all constituting one 



