132 Proceedings of the Royal Physical Society. 



The results of my observations are summarised in Table III. A normal 

 specimen has four genital sacs and four gonads, symmetrically placed round 

 the mouth with its four oral lobes (Fig. 1). In the thousand examples I 

 examined, 98 - 2 per cent, had the normal number of gonads. These, however, 

 were not always in perfect symmetry as to size or position or degree of 

 separation from each other. But variations in such respects are so difficult 

 to estimate that details are here given of numerical variation only. It may 

 well be that an accident at or before the time of capture, or even attempts 

 to place a specimen in the best position for observation, might, without 

 seriously damaging an individual, upset the original symmetry of its parts. 

 Of the eighteen abnormal examples in my collection, two show a decrease and 

 sixteen an increase in the number of gonads. One result is that provided the 

 functioning of each gonad remains constant, there is an increased productive- 

 ness owing to this variation ; for instead of 4000 gonads there were in my 

 collection 4020, and thus, what may be called the " coefficient of fertility " is 



4020 

 4000 ~ 1)b - 



The two examples with fewer than the normal number of gonads have 

 only five gonads between them : hence the coefficient of fertility for speci- 

 mens with a decreased number of gonads is 0'625. The sixteen individuals 

 with an increased number of gonads have a total of 87 instead of the normal 

 64, and thus the coefficient of fertility for them is l - 36. In place of the 

 normal four gonads, there are on the average 54 gonads for each of the 

 eighteen abnormal forms. Thus such variation as is found in the number of 

 gonads is markedly towards an increase beyond the normal. 



It is also seen from Table II. that no perfect correlation exists between 

 the number of gonads and the number of tentaculocysts and radial canals. 

 Of the eighteen forms with an abnormal number of gonads 



1 specimen with 2 gonads had 4 tentaculocysts. 



9 specimens „ 5 „ „ 9-11 „ 

 7 „ 6 „ „ 7-14 „ 



The Table, however, supports Browne (op. cit., p. 107), who pointed out that 

 a decrease in the number of genital sacs is usually accompanied by a 

 decrease, and an increase by an increase, in the number of tentaculocysts and 

 radial canals. Browne found a single specimen (in 3000) with a decrease in 

 the number of genital sacs and an increase in the number of tentaculocysts, 

 while f found a single specimen (out of 1000) with an increased number of 



