262 Records of the Indian Museum. [Vol. Ill, 



The number of the stigmatic bands varied from 5 to 19 but the 

 most common type of capsule possessed 12 or 13 bands. It is not 

 surprising that those plants which possessed few bands should pro- 

 duce few seeds, but it is most remarkable that those capsules with 

 18 or 19 bands should also produce few or no seeds. Similar obser- 

 vations were made on other plants ; the following passage appears 

 in \"ernon's book : ' ' These experiments Professor Pearson holds to 

 illustrate a ver\' important law, namely, ' Fertility is not uniformly 

 distributed among all individuals, but for stable races there is a 

 strong tendency for the character of maximum fertility to become 

 one with the character which is the type.' " It has also been shown 

 that certain medusae which are abnormal as to the number of their 

 radial canals are less fertile than those which are normal. It is 

 l)lain that this law is an explanation or discover}^ of the means b)' 

 which a race remains normal in spite of variation, for it states that 

 normaUty can perpetuate itself with ease while abnormality does so 

 with difficult}'. This explanation seemed so feasible that I felt 

 almost sure that if the fertility of these rats was to be compared 

 with some other variable character, an interesting illustration of 

 the law would appear. The character chosen for comparison was 

 the weight of the rat. House rats throughout India are very vari- 

 able in size ; in the combined length of head and body the}- show 

 a range of variation which is as wide as forty per cent, of the mean 

 length, in weight the range is much wider even than this. The 

 difficulties of measuring a freshly killed rat are well known to any 

 one who has made the attempt ; the length of a rat varies with 

 the attitude it is made to assume, and the interval of time since its 

 death. Although the weight of a rat doubtless varies according to 

 the stage in the pregnancy reached, the contents of the stomach 

 and the state of the nutrition, a record of the weight affords the 

 best estimation of the total size which is available to us. 



The result obtained was quite unexpected. The maximum 

 fertility of rats (as measured by the number of young which they 

 produce at a birth) is not one with the character which is the type, 

 as regards size. In other words, gigantic and dwarfed rats are 

 just as fertile as common rats of average size. 



The tables I and II show the results which were obtained 

 independently at Poona and Belgaum. The same methods were 

 employed at both places. The rats were weighed immediatel}^ 

 after death by chloroform , on a spring balance which was graduated 

 to record five-gramme units. The observers had attempted to guess 

 the weight to the nearest gramme, but it was plain from the way in 

 which the fives and tens predominated among the records that such 

 guessing could not be relied on. In arranging the figures, therefore, 

 the following plan was adopted : Those numbers which ended in 

 live or less were added to the ten group next below, while those 

 which ended in six or more were added to the group above. A rat 

 shown as weighing 75 gms. is placed in the sevent^^-gramme group, 

 while another of 78 gms. is placed in the eighty group. It should 

 be mentioned that the number of young within a pregnant female 



