AGRICULTURAL BOTANY. 627 



In another paper (E. S. R., 20, p. ?,25) the author has called attention to the 

 infrequency of bud variations. A further study has afforded no reason to change 

 his opinion that practically all, if not quite all, bud variations are losses of a 

 dominant or an epistatic character allowing the appearance of a recessive or a 

 liypostatic character. The experiments on which this opinion was based in- 

 \olved investigations as to change of color, shape, depth of eyes, and habit of 

 growth. 



A series of attempts was made to produce variations through grafts, such 

 variations having been rei)orted. More than 100 attempts were made to graft 

 buds of white and colored tubers, in each case the bud being cut in half at its 

 Aortical point and two grafted buds brought together. It was found that the 

 new plant was always formed from the growing tissue of one variety. A second 

 experiment, consisting of inserting a bud cut from a tuber of a colorless variety 

 into a hole of the same size in a colored one, and vice versa, gave results which 

 indicated that there was absolutely no influence of the stock upon the plant. 

 I'rom his experience the author is led to believe that many of the so-called 

 graft hybrids were plants exactly like the stock, produced from it by unnoted 

 adventitious buds. 



In summing up his investigations the author states that the behavior of 

 variations reproduced by budding is in many ways essentially like that of 

 variations coming from seed. Of the variations found all but one have con- 

 cerned characters that mendelize in sexual reproduction, and as there is no 

 evidence of inheritance in the case of this one exception, it may be left out of 

 consideration. Considering the fact that in the large amount of material under 

 (Observation only 12 inherited variations have been observed, a rather high rate 

 (;f frequency, there is believed to be no reason to recommend asexual selection 

 as a commercial means of actual improvement, as no changes of commercial 

 value have been found. 



In conclusion the author states : 



" The classification of all of the permanent bud variations as losses of char- 

 acters shows the investigation of the possible inheritance of fluctuations in 

 composition In a new light. Since the variety used in the investigation was 

 recessive in all the characters whose behavior in sexual i-eproduction is known, 

 loss probability exists that an inherited change might take place that would 

 obscure the results on the class of variations immediately concerned. Thif 

 being the case, we may feel some confidence in a conclusion that fluctuations 

 (variations due to surrounding conditions) ax*e not inherited. Furthermore, 

 there is little doubt but that the cases of so-called disease resistance should be 

 classed as noniuherited fluctuations due to various physiological causes." 



A Mendelian interpretation of variation that is apparently continuous, 

 E. M. Easi {Amer. Nat., .',Jf (1!)10). No. 518, pp. 65-82). — Some new facts of 

 inheritance obtained fi'om pedigree cultures of yellow and white varieties of 

 maize are given, followed by a discussion of hypotheses to which an extension 

 of this class of facts leads. 



The facts have to do with the independent allelomorphic pairs which cause 

 the formation of like or similar characters in the zygote. In certain cases the 

 endosperm of the maize contains two indistinguishable, independent yellow 

 colors, although in most yellow races only one color is present. There is also 

 some evidence that there are three and possibly four independent re<l colors 

 in the pericarp, and two colors in the aleurone grains. In 15 different varieties 

 of yellow maize crossed with various white varieties, crosses were obtained 

 that gave a simple monohybrld ratio, but in other cases a dihybrid ratio was 

 the result. In certain cases there appear to be several allelomorphic pairs, 

 each of which is inherited independently of the others, and each of which is 



