52 
PACIFIC SCIENCE, VoL VIII, January, 1954 
organisms and would be expected to resemble 
one another more closely than the adults of 
different species. However, when we add the 
small differences in size and shape of nema- 
tocysts to the morphological differences of 
the hydroid stages, we find that these animals 
can be readily identified as separate species. 
The nematocysts of the medusae of these 
species, on the other hand, show greater dif- 
ferences, and again we would expect this, 
from the thesis that adults of related animals 
are less alike than their younger stages. It is, 
however, somewhat disappointing to discover 
that the medusoid and hydroid stages of a 
genus do not possess the same cnidom. Thus, 
in Prohoscidactyla we find that the hydroids 
may be characterized as possessing a cnidom 
of three types of nematocysts (macrobasic 
euryteles, microbasic euryteles, and desmo- 
nemes), whereas the medusan cnidom con- 
sists of but two types (macrobasic euryteles 
and desmonemes). This should serve as a 
warning that hydroids and their medusae may 
not be identifiable with one another on a 
basis of their nematocysts. Moreover, this 
divergence of cnidoms in the two phases of 
the life history seems to imply that the nema- 
tocysts are adaptive characters which respond 
to environmental demands, and perhaps we 
should advance with caution along the road 
of relating larger groups such as genera and 
families on a basis of their nematocysts. 
In the following discussion, in order to 
speak clearly of the arrangement of the ten- 
tacles and canals of the medusae of Probosci- 
dactyla, a convention has been adopted in 
which the first four tentacles to develop are 
called first order tentacles and their canals are 
the first order canals. The second four tentacles 
arise to the right of the first order tentacles 
and are called second order tentacles. These ten- 
tacles arise at the margin of the bell at the 
point at which the second order canals join the 
ring canal. The third order tentacles are the 
the next eight tentacles which arise at the 
ends of the branches from the first and second 
order canals. Beyond the third order the ar- 
rangement of tentacles and canals is more 
difficult to discern. On specimens with 32 
tentacles a rather fourth order of tentacles 
and canals can usually be made out, but on 
specimens with more than 32 tentacles the 
arrangement seems to have lost the regularity 
presented by the younger medusae. It has 
been noticed that, after the first order canal 
gives rise to the second order branch on its 
right and the third order branch on its left, 
it usually does not branch again and that 
most later branches arise from either the sec- 
ond and third order canals or from the rather 
diffuse fourth order. 
The following material, upon which part 
of the descriptions presented herein are based, 
has been deposited in the U. S. National 
Museum. 
1. Prohoscidactyla flavicirrata, adult medusae, 
20 specimens, U.S.N.M. 50035. 
2. P. flavicirrata, hydroids on tubes of Schizo- 
branchia and Pseudopotamilla, 6 specimens, 
U.S.N.M. 50036. 
3. P. flavicirrata, hydroid on Schizobranchia, 
1 specimen, U.S.N.M. 50037. 
4. Prohoscidactyla circumsabella, holotype, 
adult medusa, 1 specimen, U.S.N.M. 
50038. 
5. P. circumsabella, paratypes, medusae, 8 
specimens, U.S.N.M. 50039. 
6. P. circumsabella, hydroids on Pseudopota- 
milla, 11 specimens, U.S.N.M. 50040. 
7. Prohoscidactyla occidentalism medusae, 12 
specimens, U.S.N.M. 50041. 
8. P. occidentalism hydroid on Pseudopotamillam 
1 specimen, U.S.N.M. 50042. 
Prohoscidactyla flavicirrata Brandt 
THE MEDUSA: This medusa was first de- 
scribed by Brandt (1834) from the north 
Pacific and has subsequently been found near 
Japan and as far south as Puget Sound on 
the American side of the Pacific. Twenty- 
eight specimens of this medusa taken during 
August, 1949 , have been found in plankton 
samples collected in the area from the mouth 
of the Columbia River on the north to Cape 
