﻿^gg MR. J. MIERS ON THE LECYTHIDACE^. 



r 



eating the junction of the exorhiza and neorhiza, the latter truncated at its summit, and 

 ohtusely pointed towards the base ; and in the middle of the ventral side we see it 

 enlarged by a nipple-shaped protuberance, which touches the margin opposite to a smaU 

 external knob m the exorhiza, having a longitudinal slit ; the protuberance is the 

 plumule ; the knob consists of 2 minute cotyledons, a structure offering much analogy to 

 that in Bertholletia and Lecythis ; in germination this i^lumule is extended into the 

 ascending stem of the new plant while the base of the neorhiza forces its way through 

 the exorhiza, to form the root. I have not witnessed this germination ; but Martius 

 describes it, and Berg portrays it in several analytical figures under the head of Le- 

 cythis coriacea \ On comparing the details of the structure of the nucleus given above 

 (the exactness of which I can affirm with confidence) with the analytical figures and 

 description of Berg, several discrepancies will be seen. Berg confirms the position I 

 have assigned to the plumule, and the existence of an external knob, with a sht in it, 

 through which it forces its way in germination to form the ascending stem ; but there is 

 evidently a mistake in the position he assigns to the sprouting rootlet, which he places 

 in the middle of the dorsal face : this assuredly would spring from the bottom of the 

 neorhiza, the existence of which was wholly unknown to him. Berg made his analytical 

 drawings entirely from the old notes of Martins, and under the belief that the nucleus 

 was a single gigantic cotyledon, instead of a monstrous radicle, as it appears to me and 

 under which point of view the phases of its germination are best explained. Ber- 

 affirms that the inner integument, as in the testa, contains many spiral vessels ; but I 



have not been able to detect them there. 



Although the genus Couratari (Plate XXXV. b) is one of the oldest and best-known of 



the famHy, its floral structure has not yet been correctly described; the only figures hitherto 



published are those of Eichard, under a reduced size^ and of Cambassedes^ aUof which 



are incorrect, Aublet never met with the flower ; nor did Poiret, who derived all his 



details from Kichard. Berg, in his monograph of the family, has created much con- 



fusion by his thorough misunderstanding of the subject ; even when he saw a flower of a 



true Couratari, he failed to recognize it, but described and figured it, as well as its fruit, 



under the name of Lecythopsis\ while his generic character and his several analytical 



drawings, given under the name of Couratari ^ actually belong to Carmiana. Its species, 



which are not many, all form noble trees, with the foliage and general inflorescence of 



Lecythis and Couroupita : the flowers are rather large, upon short pedicels, furnishe 



with 3 deciduous bracts ; they have 6 fleshy, ovate, expanded, subequal sepals ; 6 \p' 



large oblong unguiculated petals, of which the two exterior are largest ; a large andro- 



phorura, formed of an oblong, subpatelHform basal ring, attached to the claws of e 



petals, and furnished inside with a triple series of crowded, short, clavate appendages, 



staminiferous at the apex, the stamens formed as in Lecythis ; the ligula, as broa 



Mart 



ilosed 



'±vi, i-ao. /^, wnere me two upper rows oi ugmcs ai±^jvi vu.^ i'j- • H n 



figures delineating the badly drawn shape of the seeds and the mode of their geTvam o 



Ann 



Elor. Bras. Mer 



* Mart. Flor. Bras. I. c. p. 503, tab. 75, 76. ' Ibidem, p. 506, tab. 78-82. 



