1881.] The Siphonophores. 775 
Midway between each pair of polypites on the axis of Rhizo- 
‘ physa, there will be noticed a small cluster (e), which when mag- 
nified wiil be found to have a botryoidal shape and to hang from 
the axis by a small slender pedicel. These organs are ovaries, 
and correspond with the sexual bells of the Agalma colony, 
although they never take on a bell shape as is true of the latter 
genus. How the egg is formed in these clusters, and what the 
character of the development of Rhizophysa is, no one has yet 
been able to make out with any degree of certainty. 
In recapitulation, these then are the only structures which the 
skeleton-like Rhizophysa has: an axis (6), with a terminal float 
(a), polypites, or feeding-polyps, (c), from which arise many ten- 
tacles (ad), closely set with tentacular knobs, and sexual organs 
(e) in the form of botryoidal clusters situated midway between 
each pair of feeding-polyps. 
In Rhizophysa we have one of the simplest expressions of the 
group of animals of which Agalma has been taken as a type. 
There is but one simpler related animal, and that is a form in 
Which the stem is wholly wanting, and nothing remains to indi- 
cate the affinities of the animal with the Physophoridz except the 
float. We then have a well-known Siphonophore commonly 
figured as a representative of the group and called Physalia, or 
the Portuguese man-of-war. 
In this curious animal there is no sign, whatsoever, of stem, 
swimming-bell or covering-scales, and the float is enormously 
€veloped into a bladder, which swims on the surface of the 
water, and acts ina way asa sail, to the spread of surface in which 
a raised crest also contributes. The colony of individuals is clus- 
tered on its under side, and in that position is borne along through 
the water. Physalia is, in some respects, the simplest possible 
form of Physophore and most distantly removed from the type, 
Agalma. Its close relation to Rhizophysa indicates that it is a 
true member of the group and not closely related to the floating 
hydroids Velella and Porpita, which I have already followed 
McCrady in separating from the Siphonophores.' 
The genus Physophora or the “ float-bearer,” which has given 
the name of Physophoridz to the group, is not perfectly normal, 
and differs in some respects from the type Agalma. Physophora 
has never been found in American waters, although quite com- 
"Bull. Mus. Comp, Zool., Vol. v1, No. 7. 
