232 I find thr t the only snecies of those enumerated in the 



first edition of his Species plentsrum thrt is represented in the 

 Linne?n Herbf^riura is M, tenuifoliun, v/hich has the petals free 

 to the b.-^se, five stifrraas, and an inferior five-celled ovary. 

 But in his Hortus clif fortisnus Linne enumerated tv/enty-nine spec- 

 ies v/hich he probably sevr, although the only species preserved 

 in the tortus Guff ortianus herbarium at the British ^-u-seuia are 

 -', tenuifolium, I', pomeridianun, and these merely repeat the 

 story, for five of them have eight to seventeen stigmas and cells 

 to the ovary end four have a short tube to the corolla and a 

 partly superior ovary v.lth four to five stigmas and cells, and 

 tv;enty have the petals free to the base and an inferior ovary 

 v/ith five stigmas and cells to the ovary, 'fliese of course, being 

 the same species as are described in his %edies plantarum. 

 Therefore as the bulk of the species that it is probable Linne ac- 

 tually sav'.' but may not have examined, as v.;ell as those he enumer- 

 ate':^, amounting in all to tvfenty-tv'o in number, ha.ve free petals 

 end on an inferior ovary vith five stigmas, and as these charac- 

 ters cover the majority of the species at present placed under 

 the p-enus , Inronose that they be adopted for the selection of a 

 type for the genus. 



I therefore propose to reject the above-mentioned thirteen 

 species as not strictly conforming with the chief characters of 

 the Linnean definition of the p-enus , and from the remaining 

 tventy-t-" o snecies (which all have five stigmas and an inferior 

 ovarv v/ith five cells, and the placentas on the outer wall of 

 the cells, and with the exceptions that their corolla is not mono- 

 retalous, and their fruit not fleshy, as above explained, agree 

 in all other characters with the J-innean definition of the genus) 

 to select as the type of the genus ^"esembryentheraum the first 

 ST>ec-ies described in the first edition of -Linne's %ecies plantar- 

 um th^t (excluding the character of the corolla being monopetal- 

 ous) conforms Vvdth the definition of the genus as given by Linne. 

 This sr>ecies proves to be -'. umbellatum, for the four species pre- 

 ceding it (i'. nodiflorum, ^•-. cr3'-stel linum, I'-, geniculifiorum, 

 and i-. noctifloriom) all have, as stated, a short tube to the 

 corolla and a partly superior ovary. 



So that I', umbellatum, L., may be accepted as being the type 

 of the genus because, not only do eighteen of the remaining twenty- 

 two species then known to Linne asrree with it in general habit 

 and structure, but it like^^^.se accords in these particulars v^'ith 

 the majority,'- of the species that have been described since the time 

 o:^ Linne. Therefore the accept-^nce of ^■". umbellatum, L. , as the 

 tvpe o-f* the ^enus ^-esembryanthemum v.nMl not onlj'' prevent the very 

 o-reat changes of nomenclature that would occur if one of the thir- 

 teen, re.iected species were selected in its place, but will necess- 

 itate very little alteration in the generic characters assigned 

 to the genus by Linne, the only alteration needed being the omiss- 

 ion of the words ''corolla monopetalous '' and .'dapsule fleshy,'* 

 ^nd o-*" course the addition of such details of structure as are om- 

 itted from the Linnean description. 



In the above I have t^ken into account only those species en- 

 umerated by Linne in the first edition of his -Species plantarum 

 becaus they f^lone are concerned in the founing of the genus. Yet 

 if ■'•'6 examine the fortir-f^ive species enumerated in the second edi- 

 tion of that^worl", and the two others described in the tenth edi- 

 tion of his Systema naturae known to him, the conclusion above 

 arived at is in po v^ay effected. 



^-he rejected species I em referring to other genera. 



