BLASTO&ENIC VARIATIONS. 173 



with a dark aberrant form or sport of this species, 

 Aglia lugens, which had been interbred for two genera- 

 tions. In 1889 some of the lug ens ? both male and 

 female, obtained from this cross, were crossed with 

 normal tau specimens. About half the offspring ob- 

 tained resembled one parent and half the other, inter- 

 mediate forms being absent. On breeding some of the 

 1889 $ and $ lug ens together, however, their off- 

 spring consisted of about 36 per cent, of taUj and 64 per 

 cent, of lugens forms. In 1890 some of these lugens 

 were bred together, and their offspring consisted almost 

 entirely of lugens, only 11 per cent, being of the tau 

 form. In this latter case, therefore, both parents and 

 all four grandparents were lug ens; in the 1889 off- 

 spring, both parents but only two grandparents, and in 

 the 1888 offspring only one parent and two grand- 

 parents. 



If sports be of an essentially different nature to nor- 

 mal variations, as the somewhat insufficient evidence 

 available may perhaps be taken to indicate, how is it 

 that they arise ? Apparently they occur spontaneously, 

 but doubtless some exciting cause must exist, invisible 

 though it may be. The artificial production of mon- 

 sters seems to throw some light on the subject, and 

 hence a brief reference to them may be made. These 

 monsters or malformations probably differ from sports 

 only in degree, and not in kind. Hence, if the means 

 adopted for their artificial production are such as may 

 occur under natural conditions, it seems possible, and 

 even probable, that sports themselves may owe their 

 origin to similar agencies. For instance, Dareste, as 

 long ago as 1877, described numerous experiments on 



