1917] AGRICULTURAL BOTANY. 225 



and a tendency to excessive boll shedding amounting to complete sterility in 

 some plants. He has succeeded in growing a few plants from self-fertilized 

 seeds, all of which proved to be rogues, while in case of seeds not self-fertilized 

 the majority of the plants were rogues. 



A large number of crosses were made between rogues and Sea Island cotton, 

 the characters of the latter being completely dominant to those of the former. 

 Later, one of the rogues threw off some of the rogue characters. It is thought 

 that the rogue may be considered as a retrogressive mutation due to the loss 

 of a single factor, the deficiency in the proportion of rogues being explained as 

 due to their weakness and liability to early attack by angular leaf spot disease 

 and by mole crickets. Further work in this connection is considered necessar.v. 



On the partial sterility of Nicotiana hybrids made with N. sylvestris as 

 a parent, II, T. H. Goodspeed and A. H. Ayres {Univ. Cal. Pubs. Bot., 5 (1916), 

 No. 9, pp. 273-292, pi. 1). — Having continued the investigations previously noted 

 (B. S. R. 29, p. 320), the authors state that the Fi hybrids between N. tabaciun 

 varieties and N. sylvestris produce very little pollen of normal appearance, 

 while the anther cells show, almost exclusively, shriveled, functionless grains. 

 The apparently normal Fi pollen did not germinate in its own stigmatic secre- 

 tion, in that of the parents, or in any one of a great variety of artificial germi- 

 nating fluids. The pollen of the parents germinated readily in the stigmatic 

 secretion of the Fi flowers. 



The evidence is considered to oppose the view that specific chemical sub- 

 stances play an important role in determining whether or not pollen will 

 germinate, and certain results of tests made on germination and growth reaction 

 are deemed to be the effect of the reagents upon the swelling of cell colloids. 

 The absciss-layer formation is the cause of the fall of flower and fruit, and the 

 stimulus thereto is nonfertilization. Fall of flowers and fruits can be retarded 

 by lowering the total concentration of available mineral materials or nutrients, 

 variations in individual constituents being ineffective. 



A few normally matured ovules capable of fertilization are produced in 

 Fi flowers, and a little viable seed is formed after pollination with the normal 

 pollen of the parents. Back crosses are difficult to make in the field, but plants 

 under conditions of low nutrition retained their flowers longer and back crosses 

 were usually successful, though no increase of fertilizable ovules resulted from 

 this more favorable condition for successful back crossing. Grafts between the 

 Fi hybrid and its parent N. tabacum macrophylla resulted in pollen similar to 

 that from the same plants grown on their own roots, but the flowers and fruits 

 were better retained in the former case. 



On the partial sterility of Nicotiana hybrids made with N. sylvestris as 

 a parent, III, T. H. Goodspeed and J. N. Kendall (Univ. Cal. Pubs. Bot., 5 

 (1916), No. 10, pp. 293-299).— Continuing the work noted above, the present 

 paper gives the results of experiments to ascertain the mode of abscission of 

 flowers and fruits on the Fi species of Nicotiana hybrids. These studies are 

 considered to yield further evidence regarding the relation between successful 

 pollination and fertilization on the one hand, and abscission of flowers and 

 fruits on the other. The general problem and the literature of abscission are 

 to be dealt with later in a more extended discussion. 



It is stated that in N. tabacum, N. sylvestris, and other hybrids, also in 

 N. langsdorffli, the abscisson zone is to be found at the base of the pedicel. 

 A conspicuous grooved ridge, or ring, of tissue stands out around the base of 

 the pedicel in Nicotiana species, which may indicate the position of motor tissue 

 or of a node. The position of the absciss-layer is independent of this groove, 

 being usually distant from It 5 to 7 cell layers. Abscission appears to take 

 place in any portion of the abscission zone distal to the groove, starting in the 



