1915] Goodspeed: Parthenogenesis, Parthenocarpy, Phenospermy 261 



tested by castration and mutilation experiments for the presence of 

 parthenog-enesis. On the appearance of Mrs. Thomas' original paper, 

 Professor Setchell made an additional series of experiments to confirm 

 his previous observations. This past season, in connection with the 

 experiments detailed above on ''Nic. tabacciim Cuba," I made over 

 four hundred castrations of flowers of N. Tabacum var. macrophylla, 

 N. angustifolia, K. Tabacum "Maryland" and N. sylvestris. The 

 number of plants in each case was twenty-five and the castrations were 

 made just as the plants were "going out of their first bloom," which, 

 according to Mrs. Thomas, is the most favorable period for the pro- 

 duction of parthenogenetic seed. The plants were also so thoroughly 

 cut back that the castrated flowers were the only ones left on the 

 plant ; all maturing and mature seed capsules were also removed. This 

 treatment, as noted by Howard (1913 p. 41) and in our cultures 

 also, usually induces heav}^ and rapid seed production. In addition, 

 the occurrence of parthenogenesis has continually been tested for 

 in connection with the making of a very considerable number of 

 hybrids between many of the species and varieties listed below. 

 Whenever a castration, preliminary to cross-pollination, is made, an- 

 other bud or two on a difi'erent part of the plant is also castrated and 

 left unpollinated at the time of making the cross. We may further 

 add to this summary, as being significant, the hybrids between N. 

 sylvestris and various N. Tahacum varieties. The F^ hybrids are 

 completely self-sterile since, in our cultures of them, no normal pollen 

 is produced. Over 500 bag'gings have been made of F^ flowers and 

 a great variety of efl^orts have been made to stimulate seed production 

 according to the methods described by Wellington (1913). In no 

 case did parthenocarpy or parthenogenesis result. It is important to 

 note in this connection that ]\Irs. Thomas reports an F^ hybrid between 

 ''A^. sylvestris and N. tabaccum Cuba" and an F,, of the cross "N. 

 sylvestris by N. affinis" as producing parthenogenetic seed. I have 

 never been able to make the latter hybrid successfully and have yet to 

 see a strictly fertile F^ species hybrid involving N. sylvestris as a 

 parent. In our cultures the many hybrids that have been made have 

 never given any evidence of the production of parthenog'enetic or 

 apogamous seed in the sense that hybrids have bred true to the maternal 

 parent from F^ on through later generations. 



