THE AMERICAN NATUBALIST [Vol. LV 



have periods 1 and 2 run over the same number of days. 

 Among the animals there should be included several adults which 

 had been brought to condition of underweight in the preliminary- 

 period with vitamine-free foods. This would answer any possible 

 question concerning the effect of the intermittent growth im- 

 pulse. It is hoped that others may be able to carry on this 



These experiments were performed und^r a grant from the 

 Committee of the Permanent Charity Fund, Incorporated. 



R. R. Renshaw 



Harvard Medical School, Boston 



INTERFERENCE IN PRIMULA StNENSIS 

 Outside of Drasophila, the only data bearing on the question 

 of interference of crossing over are those- which I reported in a 

 paper on linkage in Primula sinensis.^ The phenomenon of in- 

 terference — our knowledge of which in DrosopMla dates from 

 crosses made by Sturtevant and analyses made by Muller in 

 1912— consists of the fact that the occurrence of a crossing over 

 in one region of a chromosome reduces the chances for the oc- 

 currence, in that cell, of another crossing over in a different 

 region of the same chromosome; thus there is a smaller number 

 of double crossovers than would otherwise be expected. The 

 amount of interference is expressed by Muller 's index called 

 "coincidence," which is the ratio of the proportion of double 

 crossovers actually observed in the experiment to the proportion 

 of double crossovers which would have been expected to occur on 

 the assumption that crossings-over in the two regions were inde- 

 pendent of each other; the latter, or "expected" proportion of 

 double crossovers is obtained by simply multiplying together 

 the proportion of crossovers in one region by the proportion of 

 crossovers in the other region. As I stated in my paper on 

 Primula, a calculation based upon my total results could not be 

 sufficiently reliable to decide the question of whether or not in- 

 terference existed in Primula. This was on account of an uncer- 

 tainty in the classification ; I now find, however, that a calcula- 

 tion based upon, a selected group of the plants, in which the un- 

 certainty does not exist, is sufficient to decide the question in the 

 affirmative — contrary to my earlier conclusion. 



Three pairs of genes were involved in the Primula crosses — 

 1 Altenburg, E., 1916, Linkage in Primula sinensis. Genetics, 1 : 354-366. 



