170 Davis. 



most of its characters that de Vries presents a formula 6 X m = m. 

 The reciprocal of this cross, muricata pollinated by biennis also gives 

 a uniform Fi generation, in most of its characters so strongly patrocli- 

 nous that the situation is expressed by de Vries in the formula m X 

 b = b. The double reciprocals present two possible arrangeineuts 

 (muricata X biennis) X (biennis X muricata), and (biennis X muricata) X 

 (muricata X biennis). DE Yeies states that the double reciprocal crosses 

 give plants with characters almost wholly of the parent whose position 

 is peripheral in the formula. The characters of the parent in the 

 center of the formula almost wholly disappear in the double reci- 

 procal hybrids. Shortly expressed the results, according to de Vries, 

 justify the equations {m X 6) X (b X m) = m, and (6 X »n) X (m Xb) =^b. 

 On the appearance of the preliminar}' paper of de Vries in: 1911 

 I made plans to repeat his crosses between biennis and muricata and 

 in my studies with other species of Oenothera to grow reciprocals and 

 double reciprocals that further data might be available on the interesting 

 problems presented. I can now report on reciprocal crosses involving 

 four different sets of parents two sets of wliich have not been studied 

 by de Vries. They are the reciprocal crosses between (I) biennis and 

 muricata, (2) biennis and franciseana, (3) biennis and grandiflora, and 

 (4) biennis and gigas. Discussion of the problems involved will be found 

 in the separate accounts of the hybrids. 



(1) Reciprocal Crosses between Oenothera biennis Linnaeus and 

 O. muricata Linnaeus. 



The parents species of these crosses have been grown in my experi- 

 mental garden for the past three years and originally came to me through 

 seeds kindly supplied by Prof, de Vries. They are then the types which 



de Vries employed in his ex- 



[ J r\ C\ \\ periments and their hybrids have 



\/ ^'^ ^'^ ^'^ agreed in essentials with the 

 \,aiinv5 T^i^l, \,xm maTvtata. hybrids obtained by de Vries. 

 Fig. 1. Cotyledons. Showing the hybrids of Couscquentlv my studies in gen- 

 6«enn,> and ,»«rK-«te in certain respects inter- ^^.^j ^^^^.^^; ^j^^ ^^^^^ v^X>on^<l 

 mediate between the parents although in the 



main patroclinous. '^Y ^E \ RIES although I am in- 



clined to lay a different emphasis 

 on the conclusions or interpretations based upon the results. 



Oenothera biennis of Holland is a species so far known to the 

 American botanists only thniui:li plants or their descendants grown from 



