GENERA CAMELLIA AND THEA. 339 



Camellia and Thea in the ' Transactions of the Horticultural Society of London,' vol. vii- 

 p. 519. The essential characters which he assigns to them respectively are as follows : — 



Camellia, Linn. 

 Flowers axillary, sessile. Calyx inferior, of 7, 8, 

 or more deciduous imbricated scales, the inner 

 ones the largest. Corolla of 5 petals. Styles 

 united nearly their whole length. Capsule fur- 

 rowed, having as many cells as furrows, and one 

 or two seeds in each. 



Thea, Kaempf. 

 Flowers axillary, stalked. Calyx inferior, of 5 

 deeply divided permanent roundish segments. 

 Corolla of 5, 6, or 9 petals. Styles cohering at 

 the base, dividing towards the point into three 

 distinct bodies. Capsule 3-lobed, 3-celled. 

 Seeds solitary. 



Here another characteristic feature of the two genera is pointed out, viz. that in 

 Camellia the flowers are always sessile, while in Thea they are stalked, though it can 

 hardly be admitted in the essential characters of the two genera. The styles in Camellia 

 are said to be united nearly their whole length : that is certainly the case in all those 

 species with which Booth was acquainted in 1830 ; but in my Camellia Hongkongensis 

 they are entirely free. Applying his generic character to the Camelliacece known at that 

 time. Booth followed Lindley's suggestion, and transferred Camellia euryoides, Lindl., to 

 Thea. According to that rule, he ought to have done the same with Camellia maliflora, 

 Lindl. The fact that the latter was only accessible to him in a double state may partly 

 be considered the reason why he failed to do so. 



The last author who has attempted to frame a character for the two genera under con- 

 sideration is Choisy, in one of the most elaborate papers that have as yet appeared on the 

 whole order of Ternstroemiacece. It was published in 1855 in the ' Transactions of the 

 Natural History Society of Geneva.' He defines Camellia and Thea thus : — 



Thea, Kaempf. 

 Calyce simple, a un seul rang. Capsule dehiscente 

 dans tout sa hauteur. Cloisons non separees 

 du placenta central. 



Camellia, Linn. 

 Calyce a estivation imbricative, sepales sur plu- 

 sieurs rangs se recouvrant comme des tuiles. 

 Capsule semi-dehiscente ; cloisons separees en 

 haut du placenta central. 



It does not appear to me that we are justified in saying the capsule of Camellia is semi- 

 dehiscent, and its septa do not separate from the central placenta, and that the capsule of 

 Thea is quite dehiscent, and the septa do not separate from the central placenta. The 

 fact is, that in Thea the septa do at one time separate from the central placenta, and at 

 another they do not. In normally developed capsules they generally do separate. Hence 

 it follows that the dehiscence or semi-dehiscence cannot be regarded as a character of 

 generic distinction between the two genera. 



There is, however, one point which has been quite overlooked by all who have written 

 on this subject, and which seems the more important as it is not one of degree. On 

 examining the stamens, I was struck with the fact that there was in all Camellias and 

 Tlieas, besides the outer series of monadelphous stamens of indefinite nvunber, an inner 

 series of free stamens, definite in number. With a little practice the outer series of these 

 stamens is easily removed, and the inner one, closely sm-rounding the ovary, laid bare. In 

 Camellia I found them to be double in number to that of the normal number of petals, 

 and in Thea equal to it. I may add, that in the latter respect Thea agreed with -n hat I 



2t2 



