818 EXPERIMENT STATION RECORD. [Vol.35 



AGRICULTURAL BOTAFY. 



Hybrids of the genus Epilobium, R. Holden (Amer. Nat., 50 {1916), No. 592, 

 pp. 243-247, figs. 4)- — The author has extended the investigations of Jeffrey 

 (E. S. R., 31, p. 823; 32, p. 521) on the relation of hybrid infertility to defective 

 or abnormal development of the gametic elements. He now reports on some 

 observations of two sections of the genus Epilobium, Chamaenerion and Epi- 

 lobium proper, the latter having typically regular flowers and spores persistent 

 as tetrads, the former habitually showing considerable irregularity and even 

 absence of protoplasm in some of the spores, a condition supposed to indicate 

 impurity. 



Investigation of North American forms agreed with the expectation as above 

 noted, but English specimens showed abortive spores not only in E. montanum, 

 E. parviflorum, and E. hirsutum, but also in E. angustifoUum. This fact, at 

 first apparently contrary to previous observations, appeared on further investi- 

 gation to be due to the presence of two varieties of E. angustifoUum in England, 

 E. macrocarpum and E. hr achy car pum, and to the fact that wherever these are 

 found growing together cases of partially abortive spores are also found, indi- 

 cating a hybrid origin of such plants. This state of affairs is said to exist not 

 only in England but probably also in Europe, Asia, and western North America, 

 where both varieties are known to coexist ; while the spores are all normal in 

 localities where but one variety exists. Chamaenerion is therefore regarded as 

 affording confirmation instead of refutation of the value of abortive pollen 

 grains as a test of hybridization. 



The genetic behavior of the hybrid Primula kewensis and its allies, 

 Caroline Pellew and Florence M. Durham {Jour. Genetics, 5 (1916), No. 3, 

 pp. 159-182, pis. 7). — The authors have made a progress report on their study 

 of the genetic behavior of P. kewensis (known to have originated in 1900 as a 

 hybrid between P. floribunda and P. verticillata) , its parents, numerous deriva- 

 tives, and genetic combinations. The experiments consisted in breeding the 

 various forms in large numbers and in making all cross-fertilizations possible 

 among them. The results are given in considerable detail. 



It is stated that from the cross P. verticillata by P. floribunda and the recip- 

 rocal, and, in fact, between any two forms, plants representing the female parent 

 are usually obtained. The suggestion is considered as evident that this is a 

 case of actual parthenogenesis or else of monolepsis (in which ovule develop- 

 ment requires the stimulus of fertilization, but in which, however, the pollen 

 grain really makes no genetic contribution to the final product). These mater- 

 nal hybrids rarely show evidence of segregation when used for crossing and 

 usually breed true to type on self-fertilization. It is stated that under most 

 stringent tests positive evidence of true parthenogenesis has been obtained, but 

 the evidence as a whole favors the theory that while the ovules can develop 

 without fertilization they more commonly develop in consequence of that 

 stimulus. 



The offspring of the tetraploid P. ketcensis, both on selfing and crossing, gen- 

 erally reproduced the female type simply, but a single diploid individual has 

 been produced from this form. This has exhibited segregation of an almost 

 normal kind and has given new forms representing combinations of P. flori- 

 bunda and P. verticillata, the seeds of which generally germinate freely in con- 

 trast with those commonly obtained from P. kewensis. Among these new forms 

 are many shades of yellow not previously known in these plants. The fac- 

 torial relations are still under investigation. The diploid plant has been crossed 

 with the tetraploid P. kewensis, the results being plants intermediate between 

 P. floribunda and P. kewensis and a few resembling P. floribunda. 



