20 ILLUSTRATIONS OF INDIAN BOTANY. 



2 or 4-celled, sometimes but rarely 3-celled. I have from Ceylon what appears to he a small 

 flowered variety of B. aculangula^ it may however be ajjistinct species, but that, my specimens 

 scarcely enable me to determine. 



The number of cells of the ovary of Careya^ like that of Barringtoniay seems to vary, as in 

 the accompanying figure, which, though not prepared under my eye, I have every reason to 

 believe a correct representation of the specimen, five are shown though four is the normal 

 number. 



EXPLANATION OF PLATES 99 & 100. 



CareyaArlorea^ Roxb.— 1, Flowering branch, slightly imbedded in soft, loo e, cellular matter, which occupies 



reduced, in size. 



the whole interior of the fruit. 



2-3. Fasciculi of stamens, back and front views. U. A seed, detached. 



4. Anthers, back and front views. 10. The same cut. transversely, the central dark spot 



6. Ovary, cut vertically. representing the form and position of the embryo in the 



6. Ovary cut transversely, but showing 5 in place of midst of a copious albumen. 



4-ceUs, the usual number. H. The same, cut longitudinally— both from young 



7. A cluster of full grown fruit. specimens. 



8. One of them cut transversely, showing the seed 12 & 13. The same parts, as seen in ripe seed. 



Sub-order Philadelphe^. 



Much 



of plants should occupy in the Botanical system. 



small tribe 



Jussieu originally referred them to Myrtaceae, Mr. D. Don, afterwards removed them 

 thence and constituted them a distinct order, assigning as their nearest affinities ^axifmgeae, 

 DeCandole so far adopted this view as to indicate Hydvavgeae,^ tribe of ^axifraaeaeU%\\x^xx 

 nearest allies but notwithstanding, stations P hiladetpheae n^xt Myrtaceae: Z)eS;r/a however 



^''VZ ^x^aT^c""'' ^'"^^H ?"^.'^^-«ott both adopt the order, both place it near Myrta- 

 ceae and both differ from DeCandolle in considering Deulzia one of its genera. Bartliiig on 

 the other hand takes a totally different view and places the genera referred here among his Ona- 

 gmr^ae a station, m my opinion, most erroneous since it virtually reduces that order, by depriv- 

 ing It of Its only really good character, the binary arrangement of all its parts, "every part of 



ai pZad^faTZ'^^t" "'"^''P^' '^ '^"" " "^^' ^^^^^^^^ ^^--^^^^ but lost on the union 

 oi P hiLadelpheae ^hM:\i have a qumary arrangement. i)e«72m, referred here bv Lindlev Ar- 



no , and Meisner, is excluded by D.C. and Bartling. Meisne'r, differing fioni these kading 



X^aS 'r'i^lnc it to'tr ' arrangement and replaces, as I have here done, the t be among 

 JMyitaceae, raising it to the rank of a suborder. I have the more readily followed Meisne?, 

 partly because I do no clearly see on what point, with the exception of albuminrs seed a/ood 

 distmction can be established, but principallv because it affords rr> a .1 aiDuminousseea, agooa 

 noticing an Indian order, whi^h, fiL il^^^^^^^^^^^ 

 mitted into our Prodromus, would otherwise have been passed over ^ 



Whether or not I am correct in considering this a sub-ordpr rafKo,. fK. j • 



tion which may be left for abler Botanists, andean oing better oonortuni^^^^^^^^^^ '' ' Tl' 



to determinp Tn tho rm^ari t\r..^ T « i^ r : J,^ ,& "^^''^^^ opportunities fonts examination, 

 10 aeTermine. in the mean time, I copy Dr. Lmdley's character of the order u^ ^rivpn in the 

 second edition of his natural system. oraer, ds given in uie 



"Calyx superior, with a persistent limb, having from 4 to 10 r]iv,-c;«,,o -n » i n »«. 

 with the segments of the calyx, and equal to them In nZlL w kh a 'nn .* . ? • . ""l' 

 vation. Stamens indefinite, arising in'l or 2 rows fiom throHfi Vf reTatx'^s"'''' "' " 

 distinct, or consolidated ir.to one, stigmas several ra.^c. i-. u r • r • ^^^yx- ^^ 

 cells, manv-seedfid. SppH« ^..^ff ' 5 ' u„il '^'^^- ^.^P^"'^ ^alf inferior, with frc 



Styles either 

 from 4 to 10- 



cells. many-seeded. Seeds scobiform" subulate smooth T: i ^"r'°^',^»th from 4 to 10- 

 an angular placenta ; aril ? loose membrant ' A Ibume 7l" e'h""'" '' 'h '^^k "f" 

 long as the albumen, cotyledons oval obtuse flaitlh r i . ^^ ^rtihv^o inverted, about as 

 rior, straight obtuse' slrubs. ^^^^^^^t^^^ ^^^^^ ^T' 



Peduncles ax ary or terminal in t^icho^nm^.,a "Pposne, toothed, without dots or stipules. 



times a little scurfy." ' trichotomous cymes. Flowers always white. Fruit some- 



