ON VISIBLE AND INVISIBLE CHARACTERS IN SILKWORMS. 133 



6. First Report on the Inheritance of Visible and Invisible 

 Characters in Silkworms. Bv Miss Maude L. Cleg- 

 horn, F.Z.S, F.L.S, F.E.S. " 



[Received December 3, 1917 : Read March 9, 1918.] 



The mulberry silkworm races of Bengal are all with the excep- 

 tion of one [Bombyx textor) multivoltine, but their cocoons are 

 not so good as those of the European or Japanese races, and 

 therefore when the question of reviving the Indian silk industry 

 arose, the following suggestions were put forward, viz. : that 

 endeavours should be made to obtain an improved multivoltine 

 race by crossing an indigenous variety with a European one, and 

 that our knowledge of hybridization according to Mendel's law 

 should be utilised to help in the process. 



With these suggestions in my mind I obtained from Italy, in 

 December 1910, some European silkworm seed (eggs of the 

 Bombyx mori) and some Nistri {Bombyx croesi) seed cocoons from 

 the Berhampore Government Nursery. 



The European seed was of the Italian -Japanese Hybrid, and 

 was the first generation of a cross between the univoltine (pro- 

 ducing one brood a year) yellow Italian (Indigeno giallo) male, 

 and the univoltine white Japanese female — pure yellow Italian 

 seed not having been available at the time. This Italian-Japanese 

 Hybrid is very hardy, being a first cross, and so I started my 

 cross-breeding experiments with it. 



The cocoons of the multivoltine Nistri silkworm weighed (with 

 the chrysalis removed) from 1 grain to 1*6 grains, whilst the 

 cocoons of the Italian- Japanese Hybrid (being of a univoltine race 

 and therefore bigger) weighed from about 2*5 grains to 4'6 grains. 

 My aim in my experiments with these two varieties of silkworms 

 has been (1) to make a multivoltine race (because though its 

 cocoons are smaller than those of the univoltine, yet they are 

 compensated for by the numerous broods produced during the 

 year), producing cocoons of about 4 grains in weight, and (2) to 

 see how far the good qualities of the univoltine varieties can be • 

 combined with the multivoltine character. 



I made two series of experiments : — 



A. A cross between the multivoltine Nistri $ and the univoltine 



Italian- Japanese Hybrid <3 . 



B. A cross between the univoltine Italian- Japanese Hybrid £ 



and multivoltine Nistri <$ . 



Experiment A was made with individuals selected usually from 

 three or four families in each generation, except in one of the 

 experiments in F 2 , which was made with a whole family. 



Experiment B was made with the complete family in each 

 generation, with the exception of F„ when only a few T worms 

 were reared out of four layings. 



The layings produced in F t by Experiment A and those 



