310 



Contributions to Indian Botany, 



[April 



The essential character by which Malvaceae are distinguished from 

 Byttneriacese is their one celled anthers, usually bursting transversely 

 across the apex. In Byttneriacese they are two celled and open out- 

 wardly. The first step therefore was to determine the natural Order 

 of my plant — it is a true Malvaceae. The next point was to ascertain 

 whether it was really the same as the one described by Klein and Will- 

 denow. A comparison of the accompanying figure, with the following 

 very succinct character by Klein transmitted with the specimens to Will- 

 denow, will, I think, leave no doubt on that head. " Calyx 5-fidus : Co- 

 rolla monopetala, 5-partita, basin tubulosa : stamina 10 monadelpha : 

 germen depressum, styli 5, longe flexuosi: capsula depressa,5-locularis, 

 5 sperma : semina reticulata trigona." Roxburgh's description corres- 

 ponds in every essential particular except the last ' semina (carpella) 

 reticulata,' as he speaks of them being " pretty smooth," which does 

 not, however, detract from his known accuracy, since it is not till near 

 maturity that they become strongly reticulated, a state in which he might 

 not have examined them, or he may not have observed it, as it is not 

 till the carpels are removed from the calyx that this beautiful structure 

 comes into view. 



Having ascertained its identity with the plant of these authors it only 

 remains to determine the genus. 



On comparing the characters above quoted with those of the genus 

 Sida, I find only one of any weight in which they differ. In Sida the 

 stamens are indefinite, here they are constantly definite, twice the num- 

 ber of the petals. Is this a sufficient ground for its separation ? 



According to the Linnean system, which attaches perhaps an undue 

 weight to the number of stamens, it might be supposed a sufficient 

 reason, and I have no doubt that that alone induced Willdenow to refer 

 it to a genus with definite stamens, though with only half the number, 

 without adverting to the great difference of structure of the anthers. 

 According to the natural system, which pays somewhat less respect to 

 characters taken from the number of stamens alone, other characters 

 being alike, the grounds for separation are not quite so satisfactory ; 

 though, for the reason above stated, it seems desirable, as much as pos- 

 sible, to curtail that already overgrown genus, on which account I pro- 

 pose keeping this plant distract, almost solely on the ground of its defi- 

 nite stamens. 



DICTYOCARPUS. 



Gen. character.— Calyx 5 cleft persistent, without an involucel. Ova- 

 rium 5 celled, with a solitary ovule in each. Stamens 10 ; filaments united 

 at the base, free above. Styles 5. Stigmas capitate. Carpels 5, 1 seeded 

 muticous, reticulated at maturity. 



