198 



The Journal of Heredity 



ties with the highest possible oil con- 

 tent, the smallest amount of" the objec- 

 tionable "acid ;" varieties adapted to 

 such waste lands as those of the sandy 

 areas of the Gulf Coast and other sandy 

 regions of the southern United States, 

 and, perhaps, the sandhills of Ne- 

 braska ; varieties having close, compact, 

 fruiting spikes with thm-walleil, spine- 

 less, "non-i)()pping" seed capsules. The 

 plants should be prolific in fruiting 

 spikes, early maturing, and l)ear nxcr a 

 long season. 'J'hc materials for pro- 

 ducing such varietal types already ex- 

 ist among the innumerable forms of 

 castor beans, the mam problem being 

 to bring them together into one or more 

 commercial varieties. 



So tar as the writer knows, no vcrv 

 serious attempt has been made to do 

 this, although some plant-brcedmg work 

 and variet}' testing has been done in 

 Algeria, British \\'cst Indies, Egypt, 

 and India. From several years experi- 

 ence, the writer has found the castor 

 oil ])lant an excellent subject for ])lant- 

 breeding work. Tiie plants are easily 

 grown, comparatively free from dis- 

 ease, the seeds remain ^•iablc for several 

 years with a \cry liigli per cent of ger- 

 mination. .\11 tlie types, even to the 

 most extre re, reaflily cross, giving rcr- 

 tile 1'^, and l""., hyl)ri(ls. Some varieties 

 when crossed, gi\e a much larger yield 

 of seed in the b^, generation, while of 

 other types, as in maize, this is not true. 

 Among the characters showing !\Ien- 

 dclian behavior are stem, foliage, and 

 seed coat color, glaucous or non-glau- 

 cous plants, "popping" or "non-po])- 

 ping" seed capsules (dehiscent or in- 

 dehiscent), types of seedcoat moiiling. 

 seed size and shape, height of i)lant. 

 comj)actness and size of fruiting spike, 

 time of maturity, certain leaf characters, 

 etc. Only a few of these characters 

 have been studied in enough detail so 

 that they may be placed on a factorird 

 basis. Such a character as indehiscent 

 capsules, when combined with thin can- 

 sular walls and other characters should 

 be esi)ecially valuable in a variety grown 

 for oil, since a large per cent of the 

 seeds are not wasted by the poppiir^^ 

 of the mature cai)sules. Alsf) labor is 

 saved, since it is not necessar\- to har- 



\ est the crop once or twice a week in 

 order to avoid loss, as is the case with 

 comnK)n popping varieties. This char- 

 acteristic is easily transferred as ip- 

 ])arently not more than two i)air of 

 factors are involved. 



PLA.XT IS MOXOKCIOfS 



As is well known, the castor l)ean 

 ])lant is monoecious, the male and fe- 

 male flowers a])i)earing on ditterent i)or- 

 tions of the same fl(.wering spil<e (see 

 b'rontispiece). I'.oth types of flowers 

 mature at al)out the same time, the male 

 llowers ])er]ia])s sliglitl\- earlier. All the 

 flowers, either male or female, of the 

 same sjiike, do not mature at the 

 same time. The projiortion of female 

 dowers destined to ])roduce mature 

 cai)sules, usually involve a week or so 

 in setting so that the seeds of one spike 

 ri]:)en at about the same time. Square 

 l)ottomed. light colored manila i)ai)er 

 l)ags are userl to cover the infloresc- 

 ence when selfed seed is desired or when 

 crosses are made. The sacks should be 

 of fair size, as it may l)e desirable to 

 kee]) the rajiidly lengthening infloresc- 

 ence covered for a considerable ])eriod. 

 until all danger of "adventitious" female 

 llowers a])])earing and contamination re- 

 sulting tliereb}-. has been eliminated. 

 .Vfter bagging, when selfed seed is de- 

 sired, the l)ags should be shaken e\ery 

 few days as the pollen matures, so as to 

 insure ])lenty of "fruit" setting. In a 

 windy country this may not be neces- 

 sary, but the writer i\ni\> it to be so in 

 his work in r>rooklyn. 



CI' xi'KAi.i.v i!Ki;i:i) TKrK 



( )bser\ations based on culture-- in the 

 breeding plots at the Brooklyn Itotanic 

 Garden show that many of the varie- 

 ties, much to my surprise ( since? the 

 castor oil i)lant is monoecious ami wind- 

 pollinated), breed true to man\- of 

 their most prominent characters imme- 

 diatel}-. Data from progeny of 

 ]ilants of different \arieties. grown 

 close together. imli.-ate that \ery 

 little cross-fertilization took ]dace 

 ( nrobablv not more tliaii 5''^ ). even 

 when conditions appeared most favor- 

 .able. This may be accounted for, 

 perhaps, by the copious supply of pol- 



