45 6 SCIENCE PROGRESS. 



previously described by Baillon under the expressive name 

 of Phylloxylon (27). It has broad, sometimes branched 

 phylloclades, bearing flowers on their margins, succeeded 

 by short, coriaceous ovoid pods. Imperfect specimens had 

 long lain undescribed in herbaria. Polycardia is a singular 

 genus of the Celastrineae, characterized by the flowers being 

 borne in clusters on the leaves. \r\ P. centralis they are 

 borne on the midrib in the middle of the upper surface of 

 the leaf; in P. Baroniaua (28) they are borne in a lateral 

 sinus ; and in P. phyllanthoides they are borne in an apical 

 sinus. In all cases they originate from the midrib of the 

 leaf. Structurally, however, the genus hardly differs from 

 Celastrus itself, to which Baillon has reduced it. Complete 

 specimens of Thouars's doubtful genus Dilobeia have been 

 collected, proving it beyond question to belong to the 

 Proteaceae. Baillon has figured it (29), and the genus 

 is apparently monotypic. The broad leaves are doubly 

 bifid on the young trees and barren branches, and singly 

 bifid on the flowering branches, with a large saucer-shaped 

 gland in the base of each sinus ; and the inconspicuous 

 flowers are dioecious. Primorphopetalum (30) is a distinct 

 new generic type, most nearly allied to Pnpatiens, and 

 described as having two small green lateral sepals and a 

 large petaloid, hood-shaped posticous one, and three petals, 

 of which the anticous one is similar to the posticous sepal, 

 and the lateral ones are orbicular and narrowed into a claw. 

 Tisonia (31) is an equally singular new type of the Bixineae. 

 It has apetalous, hermaphrodite flowers with three large se- 

 pals, which grow out into a three-winged organ enclosing the 

 fruit. Many other curiously aberrant types might be noted, 

 but those indicated are among the most interesting. Yet 

 I would not be understood as laying great stress on this 

 fact, or attaching great importance to it, because relatively 

 this feature is no more pronounced in Madagascar than it is 

 in Bourbon and the Seychelles, to say nothing of conti- 

 nental Africa. To give examples from other countries, 

 however, is beyond the scope of these articles. 



Not less interesting than the endemic genera are the 

 endemic species of some notable genera, especially in 



