THE INFLUENCE OF THE ENVIRONMENT ON THE 

 SIZE OF EXPECTED CLASSES. 



T. H. MORGAN AND SABRA COLBY TICE. 



In crosses in which rudimentary wings are involved, it has 

 been apparent, since this race was first bred, that the classes 

 containing rudimentary wrings often run far behind expectation. 

 The experiments made clear that the character rudimentary 

 wrings is a Mendelian recessive and is sex-linked. The defi- 

 ciencies that appeared were assigned to viability of these flies. 

 We have found meanwhile for other stocks that by breeding 

 pairs of flies in large bottles, with an abundance of food, kept in 

 good condition, there was a very marked increase in number of 

 those classes that are deficient in number if many flies are bred 

 in small bottles, or even in large bottles if so many parents are 

 used that crowding of the larvae takes place. It was deter- 

 mined, therefore, to repeat the experiments with rudimentary 

 wings under the most favorable conditions that our experience 

 had made known to us. In order to avoid the possible criticism 

 that the stock might have changed, a control culture en masse 

 was again made in w r hich crowding took place. 



Results similar to these with rudimentary wings had also come 

 up in crosses in which a new mutant, "strap wing" was involved. 

 This factor is not sex linked, but belongs to our second group. 

 Similar experiments were carried out with this stock. 



THE VIABILITY OF THE RUDIMENTARY WINGED RACE. 



In a paper in Science, 191 1, 1 an F 2 count is given in which 

 5,850 long-winged flies (cf and 9) and 83 rudimentary-winged 

 males are recorded. The expectation is that of these 5,850 flies 

 one third should be long-winged males, or 1,950. This number 

 is also the expectation for the rudimentary -winged males. Instead 

 of 1,950 there are only 83 males or 1/23 the expected number. 



1 Science, Vol. XXXIII., March 31, 1911. 



213 



