Eeactions and Peoducts in Interspecific Ceosses 161 



with oblongata dominant. 



The third F^ group in the primary cross of these species shows dominance of 

 oblongata, with relatively few complications. None of them is pure, all that 

 have been tested being the heterozygous, but in varying degrees and directions. 

 From the third F^ group many lines of homozygous-acting combinations have 

 been derived, of which the most interesting are those that involve the form deter- 

 miners, and a duplicate series shown for the first group is found, showing that 

 the form-determiner is dissociated and interchanged in all F^ interactions. The 

 series, however, gives nothing in action or principle that diifers from that seen 

 in other aspects of this species cross. 



One of the chief points of interest in the crossing of these species is in the rela- 

 tion of the two gametic systems in F^. In F^ there appears an equivalence and 

 blending of the two when the cross is not complicated by certain agents within 

 and without, whose action has been analyzed. As shown in table 17, the F^ prod- 

 uct under the conditions of the laboratory quarters is highly diverse, even when 

 the parent lines have been fully accustomed to these conditions for generations. 

 There is, of course, not constancy in this environic complex, but rhythmic regu- 

 larity of conditions which diurnal changes produce in the organisms — possible 

 differences in physical condition that may be in some measure productive of the 

 complex Fi array and the amount of the dissociation of the gametic systems 

 shown. That the conditions of the medium do have something to do with the 

 complexity of F^ is shown by the crossing of these species under the mean condi- 

 tions of the breeding-quarters, in special apparatus, but constant, when there is 

 produced in F^ only the blended intermediate type with uniformity in the entire 

 array in all characters ; but in breeding these are all strictly heterozygous, and in 

 F2 give heterogeneous arrays, depending upon the conditions present in the 

 medium. 



The complete analysis of this cross would require many times the facilities 

 that I have had for controlled conditions, but enough is given concerning the 

 cross of these two species to show that the complexity of the products of the 

 cross are dependent upon the condition of the materials and the oscillation of 

 factors present in the medium. The crossing of these two should first be 

 worked out under constant conditions, and then the role of departures and 

 changes in the medium could be determined. The cross shows how, in the ever- 

 changing conditions in any natural habitat there would be produced many pos- 

 sible combinations and rearrangements, giving many distinct lines of descent. 

 It is this operation in nature that has much of interest, because of its possible 

 bearing upon the production of heterogeneity in nature, and also because the 

 same two species have in nature been utilized in experiments in the production 

 of mutating stem-forms. 



Crossing of these species under the usual conditions of life produces what 

 seems to be an extensive disruption of both of the gametic systems, which is in 

 experience even more complicated than here indicated. Not only are the unim- 

 portant characters of color and pattern dissociated and interchanged, in differ- 

 ing aggregations, but habits, food-relations and growth are involved. As a 

 result, hardly any two cultures are alike, and derivation of homozygous-acting 

 lines is often attended with endless difiBculties; nevertheless, at no point has 

 there been anything which on test proved not to be regular in its action and con- 

 stitution. I v/ish that it were possible to present at this time even an approxi- 



