DR. M. T. MASTERS OK A DOrBLE-rLOAVERED ORCnTS. 351 



having small, tubular spurs (PI. X. fig. 5) ; they had, moreover, 

 flower-buds in their axils : thus, in this third series, there was an 

 attempt to form an anther and at the same time nectaries. The 

 fourth whorl was made up of three or, sometimes, of only two 

 sepahs (PI. XL t. 4,4,4) : if three in number, they alternated in 

 position with the segments of the preceding series ; if two only 

 were present, they were so placed as to show that it was the 

 posterior one that was deficient: an axiUary bud was present in 

 all cases. The fifth whorl, in the more perfect flowers, consisted 

 also of three pieces (PI. XI. i. 5, 5, 5), like the others; and all of 

 them were provided with axillary buds. Sometimes this whorl 

 ■was represented only by a single piece, placed anteriorly. The 

 sixth verticil was composed of three segments, but generally of 

 only two, placed laterally (PL XL i. 6, 6). The seventh circle was 

 rarely complete, being generally reduced to a single sepal with 

 the trace of a bud in its axil, and placed in the fore part of the 

 flower (PI. XL i. 7). 



The centre of the flower, within this seventh whorl, was in 



so imperfect a condition that its parts could not be satisfactorily 

 dissected. 



Before proceeding to the description of the axillary buds, it 

 may be well to inquire whether these seven w^horls of the primary 

 flower, just described, are parts of one and the same flower, or 

 whether they are representatives of a series of imperfect flowers, 

 produced, one above another, by median prolification. To this end 

 I must refer to the views held by Eobert Brown and Darwin on 

 the theoretical construction of Orchidaceous flowers*. Assuming 



e correctness of those views, there should be in a perfect Orchid 

 fifteen parts, arranged in five whorls (PL XL ii.). 



^ow, in the flowers which form the subject of this note, the two 

 outermost whorls belong to one and the same blossom, of which 



ey formed in fact the calyx and corolla. The third whorl pro- 

 bably belonged to the same flower, inasmuch as it presented traces 

 of a stamen corresponding precisely to tliat marked A 1 on Dar- 

 win's diagram, but differing from the stamen usually found in 

 this situation in being perfectly free and distinct ; indeed there 

 ■v\aa no style for it to adhere to; the two lip-like organs of the 

 same set correspond to A 2 and A 3 (PI. XL ir.), which are usually 

 Confluent with the labellum, but which are in this case distinct and 

 ^ore leaf-like in appearance than staminal. The fourth whorl 

 would correspond to stamens a 1, a 2, « 3 (PI. XL it.); and it may 



th 



th 



m 



Darwin, 'Fertilization of Orcliids,' p. 292. 



