148 ON VARIABILITY AND ADAPTATION 



tube regain the power of pigment production. This may be 

 in two or three days, or, again, only after several transplantations 

 at the end of two or three weeks ; and when we remember that a 

 bacillus divides and so forms a new generation in, on the average, 

 something considerably less than an hour, it is seen that the 

 acquired character may be impressed upon the race for some 

 hundreds of generations. 1 The more intense the alteration 

 to which the bacillus is subjected, the longer and the more 

 frequently the race is subjected to the altered temperature 

 conditions, the longer it is before there is a sign of return to the 

 normal. Translated into the terms of this theory, heat leads 

 the idioplasm to have certain side-chains, either modified or lost, 

 and this modification or loss is inherited ; but return to the 

 normal environment leads the modified idioplasm in the process 

 of growth and metabolism eventually to assume side-chains of 

 the same composition, the conditions of growth being similar to 

 those under which the species originally acquired the side-chain. 

 Yet another solution, though it is one which I do not favour, 

 is that not all of the molecules of the idioplasm undergo modifica- 

 tion ; a minority retain their original constitution and, under 

 favourable conditions, gradually come to preponderate. 



Sexual Conjugation and Inheritance. — How, next, does this 

 theory bear upon sexual conjugation and its effects, for this is 

 also met with in many unicellular organisms ? I shall not take 

 up the explanation of the development of male and female 

 characteristics ; this is too large a subject, but I would discuss 

 how fusion of the idioplasm of two individuals affects inheritance. 

 In that fusion we have a mixture of the two idioplasms, and, 

 as already pointed out, these idioplasms, a result of the varying 

 influences which have acted upon them, tend to vary in certain 

 particulars in their constitution. This fusion must be either a 

 mere admixture, so many molecules of one idioplasm becoming 

 admixed with so many molecules of the other, or a true chemical 

 combination. If we presuppose a mere mixture, we are led 

 along the lines of Weismann's theory and have to regard the 

 idioplasm of the individuals of one generation as being composed 

 of idioplasmic molecules or ids which have been passed down from 

 the long line of previous generations. I have already pointed 

 out the difficulties in which we find ourselves if we accept this 



1 See the preceding chapter upon the Variability of the Bacteria. 



