The following paper was read by D. T. MacDougal: 



SOME CYTOLOGICAL ASPECTS OF HYBRIDS 



W. A. Cannon, Columbia University, New York. 



Ijntil recently research upon hybrids has been almost entirely confined 

 to the more practical problems, such as are connected with the formation of 

 new or better sorts of plants, and students of pure science have given little 

 attention to the subject. At the present moment, however, much interest i? 

 manifested by scientists in this direction, and this awakened interest is largely 

 due to the republication by Correns and De Vries of Gregor Mendel's ex- 

 periments and results. The more recent studies have a two-fold nature. They 

 are either experimental, as the well known work of De Vries, Correns, Web- 

 ber and Tschernak, or they are cytological ; and each phase of the work has 

 been undertaken entirely independent from the other. 



What are the relations of the cytological to the experimental researches 

 of hybrids? This question can hardly be answered at this time, but the 

 possible connection between the two may be pointed out, however. 



Although it is not the purpose of this paper to set forth the conclusions 

 of the experimenters, for illustration and for reference, it will be convenient 

 to briefly present those of Mendel. 



Mendel studied the behavior of hybrid plants, not only in the first gen- 

 eration that of the immediate cross but in the following ones as well, and 

 learned (i) that in the first generation the hybrid shows the characters of 

 one parent only, but that (2) the characters which were latent in this gener- 

 ation appeared in the later ones, and in such a manner that the plants having 

 the "dormant" and the "recessive" characters bear a certain and definite ratio 

 to one another. 



In this connection it need only be said that while the conclusions of 

 Mendel have been verified by the later researches, they have been found to 

 be limited in their application to a special type of hybrid, namely, the Pisum 

 type, while the greater number of hybrids belong to other types. 



In the Pisum type the plants behave as if the bundles of inheritance 

 which were derived from the parents of the original cross were kept separate, 

 and were delivered as such to the succeeding generation, and, as if in fer- 

 tilization, these were united in all possible proportions. This has been ex- 

 pressed by the following formulae: A+A, A+a, a+A, a+a. The result is that 

 a certain per cent of the hybrids revert to the characters of the parents of 

 the original cross. Or, in other words, according to the laws of Mendel, we 

 might expect that the chromatin derived from the primitive' parents main- 



