The Bearing of Cytological Research on Heredity. 343 



himself had pointed out. It was perhaps the same difficulty that led 

 Correns (1902) and De Yries (L903), in their attempts to explain Mendel's 

 law, to treat the chromosomes as of quite secondary importance. Their 

 explanations operated almost wholly with smaller elements, of which the 

 chromosomes were supposed to consist. It is now clear, however, that only 

 when the chromosomes are taken into account do all the facts fall into line. 

 For the most recent studies in genetics have produced indubitable evidence 

 that Mendelian units are often, in fact, more or less definitely linked together 

 in groups, as they should be under the chromosome theory. 



Linkage was first clearly recognised in sex-limited or sex- linked heredity, 

 to which I have already referred. A form of linkage having no relation to- 

 sex was brought to light a little later by Correns and by Bateson and 

 Punnett in certain plants, and is now known to be of rather wide occurrence. 

 I will confine my attention mainly to the case in which both these forms of 

 linkage are now most accurately known, that of the fruit- fly, DrosopMla 

 ampelophila. In this species a very extended experimental analysis of the 

 genetic phenomena has been carried on during the past four years in the 

 laboratory of Columbia University by my colleague, Prof. T. H. Morgan, and 

 his pupils and co-workers, Sturtevant, Bridges, and many others, from the 

 investigations of whom the following results are reported. Drosophila (to- 

 paraphrase the words of Lacaze Duthiers) seems made for the experimental 

 study of genetics. It passes through a complete generation from egg to egg 

 in about twelve days. A single female not infrequently produces upwards of 

 a thousand eggs. These fortunate circumstances have made it possible, in 

 the course of four years of continuous study, to accumulate a prodigious mass 

 of data, far surpassing in extent any others thus far made known. During 

 this period these flies have given rise to more than a hundred definite 

 mutations which are inherited in accordance with Mendel's law. They are 

 of many different kinds, affecting the colour, shape, and structure of the body 

 and of the eyes, the structure of the wings, legs, antenna?, and so on. Up to 

 the present time 72 of these characters have been more or less completely 

 tested as to their behaviour when crossed with the normal or " wild " form, 

 and with one another. The all-important fact which these tests have 

 established is that the characters fall into four definite linkage-groups, of 

 which the first now includes 31 characters, the second 23, the third 17, the 

 fourth, so far, but a single one. These numbers represent of course only a 

 beginning. They steadily increase as observation continues. 



As the elaborate experimental analysis has proceeded, carried on by a 



usually transmitted in aggregates, considerable groups being derived from the same 

 progenitor" (Francis Galton, 'Natural Inheritance,' 1889). 



