500 DR. M. T. MASTERS OK THE STRtJCTUIlE OF THE 



h 



vures longitudinales, par quatre eu general/' This corresponds 

 closely with what I have observed in the living plant. A. de 

 de Jussien describes also the pollen as triangular, like that of an 

 Onagrad. It is not necessary for me to quote Jussien' s descrip- 

 tion at any greater length ; but I may add that in accordance 

 with this description he drew up revised characters for the genus, 

 and published a plate with diagrams, &c., in which the form and po- 

 sition of the several organs are represented in conformity with the 

 description. 



Not to mention other botanists who have published accounts of 

 the genus derived from the statements of De Beauvois and Jussien, 

 rather than from their own observations, such as Eobert Brown, 

 De Candolle, Desfontaines, &c., I pass on to the description 

 written by Dr. Lindley, in the Botanical Eegister, 1844, and 

 in the Grardeners' Chronicle of the same year, p. 780, in which 

 latter woodcuts are given. Dr. Lindley examined dried flowers, 

 and others in spirit, that were collected at Sierra Leone by 

 Whitfield. He gives some details of the anatomical struc- 

 ture of the wood which have not been noticed by any other 

 observer. He mentions the flowers as growing in groups of three 

 in the axils of the leaves, and alludes to Whitfield's statement 

 that, when decaying, they assume a bluish tint, which has 

 probably led to De Beauvois' s error in representing them as 

 ** almost wholly blue in their perfect state." The corolla and the 

 two whorls of the corona are described exactly as they were found 

 by me in the flowers that I have examined. The stamens are 

 mentioned as twenty in number, linear-lanceolate, with erect, ob- 

 long, two-celled anthers. Can it be that in Lindley's specimens 

 all the twenty filaments bore anthers ? This is possible, but it is 

 hardly likely they were two-celled, still less likely that they were 

 erect ; indeed, in the drawing in the Gardeners' Chronicle, where 

 half a flower is shown, there are ten one-ceUed anthers Lent down- 

 wards and inwards in the ordinary way. The fruit is as large as 

 a pomegranate, and not unlike one in appearance. The remain- 

 ing points in Dr. Lindley's description it is not necessary to 

 allude to, with the exception of pointing out an error in the 

 figure (to which the Dr. himself called attention), in which a third 

 row of the corona is made to appear between the second row and 

 the stamens. This third row (or, as Lindley calls it, foiu'th ring 

 of the corolla) has no real existence. 



In the year 1848, in the Botanical Magazine, tab. 4387, Sir 



