100 
Variation in Am^elia Aurita 
examine the radial canal system, but only the number of tentaculocysts was 
marked on skeleton diagrams. 
In the 1000 adults (Series B) there are 58 specimens (5"8 p.c.) with less than 
the normal number of tentaculocysts and the majority show that the absent tenta- 
culocyst belougs to the perradial system. 
The exact position of the absent tentaculocysts cannot be conveniently shown 
by the use of tables, but full details are given below. Specimens with less than 
the normal number of genital sacs have usually a decrease in the number of radial 
canals and tentaculocysts (Table VII.). 
The Position of the Tentaculocysts in Specimens having an Increase in 
Number. 
No less than 171 specimens (171 p.c.) show an increase in the number of 
tentaculocysts compared with 5'8 p.c. showing a decrease. This excess is one 
of the characteristic features of the variation and occurs in every series examined 
(Tables I. and VI.) and it clearly shows that the tendency is towards an increased 
number of sense-organs. It was to find out whether the perradial or the inter- 
radial series were more subject to variation that the radial canal system was 
examined, as it was the only reliable means of exactly locating the original source 
of the tentaculocysts. A tentaculocyst may be adradially, or nearly adradially, 
situated on the margin of the umbrella, but its canal may curve and join the 
perradial system or it may curve the opposite way and belong to the interradial 
system. 
I have not called any tentaculocysts or branched canals adradial, though they 
may be in some specimens adradially situated. The increase in the perradial 
system is simply a case of doubling or trebling the number of canals and tenta- 
culocysts ; in the interradial system only doubling occurs. To call such canals and 
tentaculocysts adradial would imply that the interradial canal and tentaculocyst 
were absent, and this is not the case (Diagram III.). 
The examination of the canal system shows that the increase is generally 
perradial and details are given on page 102. 
Specimens with more than the normal number of genital sacs have usually an 
increase in the number of radial canals and tentaculocysts (Table VII.). 
If a very slow and gradual change is taking place in the number of tenta- 
culocysts then the tendency is towards the establishment of a race with ten 
tentaculocysts, due to an increase of two opposite perradial tentaculocysts 
arranged thus : — 
Perradial 2.1.2.1. 
Interradial 1.1.1,1. 
If no check occurs and the vai-iation still continues in the perradial system the 
ultimate result will probably be a race with twelve tentaculocysts arranged thus : — 
