E. RuGGLEs Gates 355 



character for demonstration purposes. In using them for breeding 

 experiments, knowing they were unguarded, Shull took his own chances, 

 with unfortunate results. Though I did not record on the packet the 

 pedigree number of the plant from which these seeds came, I have 

 other reasons than the subsequent results for believing that they were 

 from the culture next to the grandiflora plants, where the opportunities 

 for crossing with that species were of course greater. So far as can be 

 learned, Shull's plants were probably derived from seeds directly from 

 this packet, in which case they would be r uhri calyx x grandiflora, F^. 

 I have already (1914) published an extended account of several years' 

 experiments with this cross and the reciprocal in F^ and F.,, as well 

 as sesquireciprocal and double reciprocal crosses. Although the same 

 strain of Oe. grandiflora was not available for all these crosses, yet 

 all the features of buds, foliage and development were the same in 

 the F., families from ruhricalyx x grandiflora and grandiflora x ruhri- 

 calyx, so that any disadvantages from this source are more theoretical 

 than real. A glance at Shull's published figures of his "ruhricalyx" 

 shows clearly by their bright red spots that they have been crossed 

 with grandiflora, and the leaf-shape also shows some effect of this cross, 

 even in the young rosettes. 



Since the differences between Shull's conclusions and mine result 

 from the fact that I was dealing with ruhricalyx while he experimented 

 with ruhricalyx x grandiflora under the misapprehension that it was 

 ruhncalyx, his conclusions obviously fall to the ground, and his 

 criticisms of my experiments on the basis of his "i-esults" are worth 

 nothing. 



Nevertheless, the origin of ruhncalyx is a matter of such signifi- 

 cance and has been so much discussed that it seems worth while to 

 restate the evidence on which rests my conclusion that ruhricalyx 

 originated from ruhri7iervis through a single unit-change. The ulti- 

 mate nature of this change will be discussed in my forthcoming book. 



In his anxiety to find a basis for doubting that ruhricalyx originated 

 as a simple Mendelian dominant, Shull does not treat the observed data 

 faii-ly. Instead of considering all the data, he omits entirely certain of 

 the, ratios which disprove his unsupported assumption that ruhricalyx 

 can have originated through the union of two independent factors for 

 red. The data, if all taken into account, are however ample to show 

 that ruhricalyx when it appeared . did actually differ from ruhrinervis 

 in a single Mendelian factor. It is, indeed, decidedly amusing to find 

 a Mendelian repudiating his own method of argument when it happens 



