M. J. C. DE MELLO A.ND ME. B. SPEUCE OK PAPATACE^. 5 



form is analogous to Siphonia (among Eupliorbiacese) which has 

 panicled racemes, each raceme terminated by a solitary fertile 

 flower, while the rest of the flowers are males. Even in the 

 $ plant of C. pajpaya it is only the terminal flower of each pe- 

 duncle that is fertile, the rest having sterile ovaries * ; so that, 

 add anthers to these ovaries (sterile and fertile) and augment the 

 number of flowers, and Form I. becomes Form II. ; let the 

 ovary of even the terminal flowers be sterile, and Form II. be- 

 comes Form III. It is almost certain that the terminal flower, 

 even when bisexual, is not fertilized by its own pollen, but by 

 that of the S flowers ; but we have no evidence on this head 

 from Senhor Mello. 



In the monograph of Papayacese in the ^ Prodromus ' (vol. xv. 

 part i.) the author, M. Alph. DeCandolle (making, no doubt, 

 the best use he could of his very incomplete materials) divides 

 the genus Carica of Linnaeus into three genera, whereof the first 

 {Papaya) has a dextrorse aestivation and a 1-celled fruit, and 

 the second {Vasconcellea^ St. Hil.) is supposed to have a sinis- 

 trorse or else a straightly-valvate aestivation in conjunction with 

 a 5-celled fruit. Out of the fifteen- species referred to Yascon- 

 cellea only three are put in the section with a contorted aesti- 

 vation; and of one of them {Carica cauliflora^ Jacq.), it is 

 said " aestivatio ignota.'' Now, as regards the first pair of dif- 

 ferences, I had a distinct recollection of having seen in a living 

 ^va\k flowers twisted in opposite directions on tJie same peduncle] 

 and I have verified it by an examination of Senhor Mello's spe- 

 cimens of C papaya. The real facts are these : the petals imbri- 

 cate each other for about one-third of their width ; in some 

 flowers it is the right margin of each .petal {e centro floris visa) 

 which is imbricated by the adjacent petal ; and the aestivation Is 

 "dextrorsum contorta;" in others, the left margin is similarly 

 imbricated, and the sestivation is '* sinistrorsum contorta." I 

 examined three peduncles of the bisexual S form. On No. 1 nearly 

 all the flowers had a straight sestivation {suhreete imhricata) ; some 

 were very slightly twisted to the left, and the solitary expanded 

 flower was twisted to the right. On No. 2, no appreciable twist- 

 ing of any of the flowers ; but all had the riyht margin of the 

 petals imbricated. On No. 3, the petals were either straight or 



* Compare this mth certain myrtles which have few- (say 5-) flowered pe- 



duncles — viz. 



terminal 



rest 



that pi'ocluces a berry, although all are equally bisexual. 



