ON THE VARIATION OF HALICLYSTUS OCTORADIATUS. 3 
margin of the umbrella, very near to a group of tentacles 
(fig. 1); the other is adradial and on the aboral side of a group 
of tentacles, a little way from the margin (fig. 3). 
Five specimens have fewer colleto-cystophors than groups of 
tentacles ; in each case one is missing. It is difficult to say 
whether the decrease is due to congenital variation or to the 
result of an injury. 
One specimen (fig. 6) with seven groups of tentacles has 
eight colleto-cystophors. An examination of the specimen 
shows that two groups of tentacles are united into one group. 
The colleto-cystophor, which has been shut from its normal 
position by the union of the two groups, is situated close to, 
and on one side of, the double group. 
Five other specimens show a similar union of groups of 
tentacles, but the colleto-cystophors correspond in number to 
the groups of tentacles. 
Mr. Hornell (1) has examined many large specimens of 
Haliclystus octoradiatus taken at Jersey. He states that 
33 per cent. show a variation either in the number of colleto- 
cystophors or in the groups of tentacles. More than half of 
these are cases in which a colleto-cystophor is absent from its 
proper position. 
Only five of the Plymouth specimens have fewer colleto- 
cystophors than groups of tentacles (about 3 per cent.). 
Mr. Hornell also examined 118 specimens taken at Jersey, 
and found 78 specimens (66 per cent.) with a capitate tentacle 
upon the apex of the colleto-cystophor. Some specimens have 
only a slight swelling at the apex with a few nematocysts, and 
others show various gradations up to a perfect capitate ten- 
tacle, just like an ordinary tentacle. The following table 
taken from Mr. Hornell’s paper gives the number of colleto- 
cystophors with a capitate tentacle in each specimen. 
