19 



Table III. 



The Chloranthal, or Corolline Scale Division. 



The very close affinity of all the families here associated will not I believe be doubted, with the exception of the Labiate and their allies, but these appear so distinctly to have the 

 Crassulacece, Callitrichacece, and Batidece, as their polypetalous forms, that the evidence of their true station being in this Division seems to be conclusive, and in fact I do not at present doubt it. 

 The affinity between the Gnetacem and Chloranthacece is stated by Mr. Bennett to be undoubted, and that between the latter and Polygonacece is even more marked, (Ann. Nat. Hist. 3rd Ser. vol. 1 . 

 p. 106,) and Piperacece also show a close approximation to Chenopodiacece, which may be regarded as substantiating the affinity of Chloranthacece to Polygonacece. See also p. 6 & 7. 



As physiological characters are of some weight, where doubt exists it may deserve notice that Crassulacece well represent Welwitschia, by growing in the most arid deserts ; and an 

 undescribed Selaginaceous plant from South Africa, collected by Dr. Wallich, has very closely the habit of Crassulacece* The monadelphous stamens of Welwitschia may be compared to those of 

 Polygonacece, especially to the monadelphous cup of Coccoloba, and also to those of Chloranthus, and it also agrees with the latter in the larger stamens being anterior. 



So large a family as Polygonacece might be expected to have their representatives in the Polypetalous Division, and although they have not been compared with Tamaricacecc, yet I believe 

 that family to be their more immediate polypetalous form. They agree in habit, in their partially monadelphous stamens, and in their tricarpous ovary, the carpels being united by their margins, 

 not otherwise materially differing except in their ovules. Their wood also is much alike, having strongly marked medullary rays. 



It is remarkable that in this Division are included nearly all the instances of wood in which there are no medullary rays, and also those in which they are so imperfect as to be reduced to minute 

 vertical cords of cellular tissue scattered irregularly through the wood, viz. Piperacece, Chenopodiacece, Amaranthacece, and Nyctaginece, described by Dr. Schleiden ; Crassulacece, by M. Brogniart ; 

 and more recently Caryophyllacece and Plumbaginece, by Prof. Oliver (Trans. Linn. Soc. vol. XXII., p. 289). To these may now be added FrankeniacecE^ Mesembryanthemec£,% Scleranthacece,% 

 and perhaps Illecebracece, \\ but in the latter instance a part of the genera will probably have them distinctly marked, except in the first two or three annual rings forming the centre of the stem. 



On this account this Division might be not inappropriately named from its variable wood, as Myrsinacece, Primulacece, and Portulacece (the two latter in some instances only), are the only 

 families included in the Scleranthal Subdivision I have examined, which have wood marked in an ordinary degree with medullary rays ; but, for the sake of uniformity, it appeared preferable to 

 name it in allusion to one of the most obvious characters of the flowers. 



This Division is very remarkable for the presence of barren stamens in the form of scales adhering to the petals, as in Caryophyllacece, Frankeniacece, Primulacece, Myrsinacece, Reaumuriacece, 

 Boraginece, and Hydrophyllacece, to which may, perhaps, be added Crassulacece, as the stamens of some genera alternately adhere to the petals in precisely the same way as the scales in 

 Caryophyllacece andFrankeniacece. Apart from the peculiarities of the wood, these corolline scales are perhaps its most remarkable general character, as they extend into the monopetalous de- 

 partment, Boraginece, and Hydrophyllacece closely corresponding in their scales with Caryophyllacece, and being as far as I have observed the only monopetalous families that do so. 



* I should not suppose it is a distinct genus, as it is very near Selago. 



f I have examined the stems of several species of Frankenia having wood with from 3 to 7 rings, and in all the wood is uninterrupted, except by vessels and perhaps traces of cellular tissue between the rings; also one of Beatsoma 



portulaccefolia ± of an inch in diameter, having 3 or 4 or perhaps more rays as they are very indistinct. 



J In the stem of a Mesembryanthemum having 7 or 8 rings of wood, the most external are as deficient of medullary rays as the most internal; the wood has very rarely slight traces of cellular tissue which are probably detached portions as 

 it were of medullary rays ; — in two other stems one having 2 rings of wood and the other 3, nothing like a medullary ray was perceptible. 



§ Scleranthus marginatus, Guss. the stem having 3 or 4 rings of wood. 



|| The stem of Paronychia Jamesii, Torr. et Gr. having 3 rings of wood, is much like that of a Frankenia. 



