442 PROCEEDINGS OF THE BRITISH ASSOCIATION. 



Mr. J. Smith exhibited to the meeting two species of undescribed shells of the 

 genera Fusus and Serpula. 



A paper was announced, but not read, by the Rev. J. Reade, " On the solid 

 Materials found in the Ashes of Plants and animals." 



Professor Lindley made some remarks " On the structure and affinities of 

 Orobanchacece." He stated that this order had been usually placed near Scro- 

 phulariacece, and in his Natural System he had included it in the Scrophulal 

 alliance. In their didynamous stamens, superior ovary, and monopetalous 

 flowers, they resembled Scrophulariacece. Schultz had placed this order near 

 Gentianacece, on account of their fruit and placentation resembling those of this 

 order. Other botanists had placed Orobanchacece near Monotropacece, on account 

 of their membranaceous foliage and parasitical habits. One important point in 

 which they differed from Scrophulariacece, was the position of their carpels, with 

 respect to the axis of inflorescence. In Orobanchacece the carpels were right and 

 left, or perpendicular to the axis, While in Scrophulariacece they were fore and 

 aft, or parallel to the axis. This pointed out another affinity with Gentianacece, 

 which had its carpels in the same position. With regard to its affinity to 

 Monotropa, there was a point which had been much overlooked by botanists, the 

 presence and absence, or large and small quantity, of albumen in the seeds of 

 plants ; he had found this a very constant character, and one of the best for 

 indicating the affinities of plants. Both Monotropacece and Orobanchacece were 

 distinguished for a minute embryo, lying in a large quantity of albumen. Mono- 

 tropacece was a polypetalous order, but its structure generally compelled botanists 

 to place it amongst monopetalous plants, near Pyrolacece and Ericaceae. He 

 remarked, by the way, that the division of plants according to the presence or 

 absence, cohesion or non-cohesion, of the petals, was very artificial, and hoped 

 that it would soon be abandoned. He thought that the affinities of Orobanchacece 

 were stronger with Monotropacece, Pyrolacece, and Gentianacese, than with any 

 other orders. The Professor then made some remarks " On the Placentation 

 of Orobanche," which he said had made him doubt the correctness of the present 

 theory of the situation of the placenta. It was generally supposed that the 

 seat of the placenta in the carpellary leaf was its margin, so that it would be 

 necessarily placed alternating with the dorsal suture or pistil. Exceptions, how- 

 ever, frequently occur, as in Parnassia, Papaveracece, &c. ; and the placenta is 

 spread over the whole surface of the carpellary leaf, or on various parts of it. 

 In the carpels of Orobanche there are evidently two placentae, but having no 

 communication with the margin of the carpellary leaf. He therefore inferred 

 that any part of the surface of the carpellary leaf might become ovulized. He 

 was borne out in his opinion by the fact that leaves which occasionally produce 

 buds, produce them from all parts of their surface, as seen in Ornithogalum, 



