CLASSIFICATION OF MEDUSZ. 87 
stages of development, presents the characters of the other. The J/edusa begins life as a 
polype; as a polype it is attached; as a polype it grasps and digests its food; as a polype it 
reproduces parts that have been removed ; and as a polype it propagates by gemmation ; the 
buds being detached from the parent as soon as they have acquired the form of the latter, 
and are capable of maintaining an independent existence. But in this condition it forms no 
ova. A new and distinct series of buds (flower-buds) is produced for this purpose; these 
buds are detached like the preceding ; they become developed into perfect Medusz, in which 
state alone they have been known until recently ; and from these Medusz are produced ova 
by a true sexual process, which are first evolved into the polypoid form,and go through the 
series of changes just enumerated.” (p. 30.) 
In this theory, proposed by Dr. Carpenter for adoption instead of that of Steenstrup, I 
can see only verbal differences. The main facts upon what Steenstrup built his proposed 
law are not denied. It is admitted that a polype may produce a Medusa by gemmation, and 
that the egg of the Medusa may produce an animal altogether different from itself, but like 
the polype which produced it. The reviewer admits an alternation of forms, but he denies 
that they are generations which alternate. Yet in the ordinary sense of the word generation, 
as here applied by Steenstrup, they must either be such or be the same individual. A father 
belongs to one generation, a son to a second, a grandson to a third—at least, this has been the 
case hitherto, as far as my knowledge goes. Surely the first polype represents one generation, 
the offspring of that polype a second, and the offspring of that offspring a third. And if so, 
the middle term here being a Medusa and not a polype, and its offspring a polype again, 
which produces a Medusa, we are warranted to speak of an alternation of dissimilar genera- 
tions. It does not affect the question, if they be regarded as individuals, whether they are 
produced by gemmation or from ova; nor whether we hold with Sars that we have not an 
alternation of animals of distinct classes, but of fixed and free animals of the same class. It 
is not the less an alternation of dissimilar generations. 
But when Dr. Carpenter says that the phenomena are simple metamorphoses not really 
so great as those which present themselves in the course of the development of any one of 
the higher organisms—“ the several parts of which depart more widely from each other, and 
from the early embryonic cell-clusters, than do the polype-buds and Medusa buds we have 
been describing”—he may mean, that the Medusz produced by gemmation are not distinct 
individuals, but parts of some one capable of maintaining a separate and independent 
existence. If so, he has certainly enunciated a new theory altogether distinct from that of 
Steenstrup, but one so opposed to ordinary notions of individuality among the lower animals, 
that few, if any, naturalists will assent to it until more fully and satisfactorily stated. 
To argue that, because “ we are not in the habit of speaking of the leaf-buds and the 
flower-buds of a plant as of two distinct generations,” we are, therefore, not to regard the 
alternations of polypes and Medusz as such, is to bring the common, popular, unscientific, 
and untrue notion of the nature of a plant into a scientific discussion on the nature of animals. 
We are not in the habit of regarding a leaf as an individual—popularly, we look upon the 
whole plant as an individual. Yet every botanist knows that it is a combination of individuals 
and if so, each successive series of buds must certainly be strictly regarded as generations. 
The first generation of a Lupine, for instance, is the pair of individuals constituting the 
cotyledons of the embryo, dissimilar from the second generation, which consists of the several 
