28 CXIX. POLYGONACEE, (J. D. Hooker) [Polygomum. 
7000 ft., from Bhotan to Kashmir (absent 1n Ceylon).—DI3TRIB. Affghanistan, Tropical 
Asia, Java, Philippines, Australia, Madagascar, Egypt, Tropical and S. Africa. 
If (following Wallich's example for most of them) I am right in bringing all the 
following forms under one species, this is a far more protean plant than P. aviculare, 
with which it is generally compared, but from which it differs entirely in habit, in the 
few-nerved stipules, and rhombic smooth nut. I have spent some days in avalyzing 
the flowers and fruit of the vast number of specimens of it at my disposal, in the vain 
attempt to discover constant even varietal characters amongst them. The utmost I 
can effect is to select eleven forms, the extremes of which differ so much from one 
another that I have described them as varieties, though there is not one of these that 
does not seeni to pass insensibly into two or more others, and I regard my work as 
provisional only. If any one is specifically distinct, it is var. Miqueliana, with its 
larger flowers and all the sepals acute; I have, however, seen no other specimens than 
Hohenacker's ; and the shape of the sepals is so variable with age in other species, that 
much importance must not be attached to them. Var. micranthema, with nuts some- 
times only 4j in. long, at first sight appears very distinct, but there are transitions of 
all its characters (nuts included) into those of the commoner forms. For the rest I 
must leave the further elucidation of this aggregate species to local botanists working | 
on fresh specimens. In the matter of the citation of published works and Herbarium 
numbers, &c., I have been very sparing, because I cannot reconcile in several cases 
Meissner's descriptions and references with the specimens of Wallieh, and with 
' other authors’ writings; and the diagnoses of the earlier authors are vague and incom- 
plete. Though described as annual, the root is so often woody, that I think it must be 
at times biennial or even perennial. 
P. plebejum proper ; branches stout or slender, stipules rather short lower inter- 
nodes often longer tham the leaves upper shorter or very short, leaves à to J in. upper 
often much shorter oblong or linear obtuse margin often recurved, flowers sunk 
amongst the stipules rarely exserted, sepals short broad rounded. P. illecebroides, 
Meissn. l. c. 94. P. cliffortioides, Meiss». l. c., and in Wall, Pl. As. Rar. iii. 62. 
P. herniarioides, Herb. Strach. & Winterb. 23. P. Roxburghii, var. brevifolia & 
pedicellaris, Meissn. l. c. 93. ' 
Common from Assam to the N.W. Provinces and on the lower hills.—This is 
intended to inelude all the forms known to me that do not better agree with the 
diagnosis of the following ones. It is a common African and Australian plant. To 
it I should refer Wallich’s 1691/1, 1691/3, 1691/4, the lower left-hand specimen, 
1691/5, except the Prome specimen, and 1691 G, I, K. - 
Var. effusa ; lax, slender, flaccid, internodes longer, leaves narrowly linear or 
dilated upwards from a very narrow base tip rounded, flowers all or most pedicelled, 
sepals short broad, P, effusum, Meissn. l. c. 93, in part.—Bengal to Bombay, 
Guzerat and N.W. India. Kumaon, Herb. Strack. & Winterb. 22.— Probably a native 
of wet places, whence its characters. Strachey and Winterbottom's specimens are. 
characteristic, but Wallich's 1691/3 from Oude appears to me to be nothing but 
conimon plebejum. 
Var. indica ; branches uniformly spreading all round from the crown and leafy, inter- 
nodes short, leaves 1—4 in. uniformly spreading horizontally linear- or obovate-oblong flat 
obtuse or apieulate, stipules very short, flowers crowded in the axils, pedicels shorter 
than the perianth or 0, sepals broad 2 outer acute, nut j, in. P. indicum (& ? P. 
linifolium), Heyne in Roth Nov. Sp. 208; Wight Ic. t. 1808; Dalz. & Gibs. Bomb. 
Fi.214. P. Roxburghii, var. a. longifolium & y. spathulatum, Meissn. J. c. 98. P. 
Dryandri, Wall. Cat. 1691 H, L.—The common Mysore and Carnatic form, also 
occurring in Bombay. When more irregular in development of leaves and branches not 
distinguishable from P. plebejum proper. Both habit and locality point to its being 
P. indieum of Roth, who describes the occasional scabridity of the end of the leaf as 
serrulation, and suspects its identity with P. plebejum. It is a common Australian 
and African form. ` ' 
Var. brevifolia; rootstock very woody, branches very stout grooved often 1-2 ft. 
dark brown, internodes of branchlets very short concealed by the leaves and stipules 
which latter show as a continuous white villous band on the underside of the branch- 
lets, leaves 3-3 in, obovate acute or apiculate flat coriaceous dark brown when dry, 
