62 BOTANICAL GAZETTE [january 



not be identified. These aberrant forms are of special interest from the stand- 

 point of variation, and it is hoped that further careful records will be obtained of 

 them and their offspring. 



Shx'll presents a continuation of his earlier studies^s on variation in O. 

 Lamarckiana and some of its mutants. On account of the unsatisfactory character 

 of height, branching, and leaf form for statistical comparisons, the studies are 

 eictended to include bud characters. The fact that aU his collections of buds were 

 made late in the season, and that there is a marked seasonal variation in the size 

 of the buds, would render somewhat doubtful his statement that the periodicity is 

 probablyftoo small to need consideration. The bud characters chosen are length and 

 thickness of ovar}-, h}-panthium, and cone, the forms studied being O. Lamarckiana, 

 O. ruhrinerois, O. gigas, and O. lata. A formidable series of variation and corre- 

 lation tables is presented, the general conclusion being that the mutants tend to 

 show greater variability than the parent form, though there is no decrease in corre- 

 lation of parts. MacDougal concludes on these premises that instead of muta- 

 tions being the cumulative results of ever-increasing fluctuation, they are the initial 

 process, of which fluctuations in the mutants, gradually decreasing to a minimum, 

 are an after-effect. The hypanthium of Oenothera, which is the most variable 

 part of the bud, is considered an example of a phylogenetically new character 

 showing increased variability. 



Miss Vail contributes taxonomic accounts of O. grandijlora, 0. Simsiana, 

 O. Oakesiana, O. pannflora, and O. muricata. O. parviftora, known in Europe 

 since 1759, has recently been found native in Maine. 



Cultures of O. grandiflora and O. biennis showed the presence in the former 

 of at least two mutants, and that under the latter there is a swarm of elementary 

 specie distributed throughout North .America. Two interesting cases of bud- 

 sports were obsen-ed, one in O. Lamar ckianaX{0. LamarckianaXO. cruciata), 

 and the other in O. atnmophila, which is suspected of being a hybrid derivative of 

 O. btenms. In both cases self-fertilized seeds of the mutant branch came true 

 to the type of the sport. 



A careful account Is given of ^UcDougal^s injection experiments. After 

 some negative results with other forms, mutants were obtained from injected 

 ovanes of O. biennis and Raimannia odorata. The solutions used were ZnSO^ 



VX.500;, *^a!,i\U3J2 11:1000 

 c^Qces were treated with a r^ 



Other inflores- 



..^..v. ...^ a. ^ucu ui rauium emanations, in the O. biennis experi- 

 ments, m addition to the usual forms, one rosette appeared which differed widely 

 from any know type, and its progeny was like it. In the case of Raimannia 

 odorata, a group of mutants appeared in the progeny of injected ovaries. It was 

 also found that "plants of the proarenv of the fiT^.t tmn^..... .-t,;.t, ..... .^^.«.ntl^ 



normal pelded seeds which gave a few at>'pic forms. 



fl 



« MAcponG.^, D. T., asdsted by Vail, A.M., Shuxl, G. H., and Small, J. K., 

 Mutants and hybrids of the Oenotheras. Carnegie Institution of Washington, Publ 

 no. 24. pp. 5;. pis. 22. fgs. 13. 1905. 



