PRODUCTION' OF NEW SPECIES. 2C)I 



breed, however, as far as my experience goes, they breed true to type, but 

 from the nature of this work and the brevity of my experience with these 

 forms, I do not care to be very positive concerning them. 



Experiments with other conditions, as, for example, hot moist, cold moist, 

 and cold dry, have also been tried with varying success.- Thus 10 males 

 and 10 females, apparently pure-pedigreed material, were subjected to a hot 

 moist condition for the first three-fifths of the reproduction cycle and kept in 

 normal conditions for the remainder. The two lots of eggs obtained num- 

 bered (A) 601, (B) 590. From (A) I got 124 imagines, as follows : of tortuosa 

 1 male, of melanicum 10 males and 13 females, and of normal decemlineata 

 45 males and 45 females. The tortuosa male (A 1) was crossed with a 

 female decemlineata, giving hybrid offspring, with decemlineata dominant, 

 and in the following generations a Mendelian separation into tortuosa, 

 hybrids, and pure decemlineata. This continued, the tortuosa breeding pure. 

 The melanicum were reared and bred true to type, as did the unmodified 

 decemlineata. Text-figure 27 shows the procedure and the results from this 

 experiment. 



This series of cultures, grouped as one experiment, shows again in a clear 

 manner that the variations arose only in those germ cells subjected to extreme 

 stimuli, while the germ cells developed and fertilized under normal conditions 

 did not show in any of these experiments the slightest trace of the production 

 of these highly divergent permanent variations. 



L. decemlineata, although most easily accessible and hardy, and not difficult 

 to breed within limits, is, however, the poorest of the entire genus for this kind 

 of work, owing to its adaptation to a wide range of environmental conditions, 

 to which it adjusts itself without any apparent change. It is necessary, there- 

 fore, to use extreme conditions to get the requisite stimuli, and unless one 

 be fully acquainted with the peculiarities of the habits and needs of these 

 beetles the experiment can end only in failure, as did my earlier efforts, and 

 not a few of the later ones. Among the tropical members of the genus are 

 many excellent species for this kind of experimental work, owing to their 

 adaptation to rather uniform conditions and to their ability to exist and 

 reproduce under changed conditions quite different from their normal 

 environment. The limits of life are about the same for all the species of 

 the genus, but decemlineata is adapted and adjusts itself to nearly three- 

 fourths of the entire range, while the tropical species are adapted to only one- 

 third or one-half the range ; hence experimentation with the tropical species 

 is far easier and much more satisfactory. 



As far as I have gone in this investigation my experiments with decem- 

 lineata show that when the animals with developing germ cells are subjected 

 to strong stimuli many of the individuals derived from these same germ cells 

 often show intense heritable modifications, whereas those not acted upon, 



