BALFOUR—NEW SPECIES OF RHODODENDRON. 301 
Of the thirty-four species named, only three are known out- 
side of China: Kh. lapponicum, Rh. nivale, and Rh. parvi- 
folium. That we are yet far from knowing all the species of 
the series we may be certain. Already we have in Edinburgh 
specimens that do not well fall into any of these described species, 
but which cannot be described because of the lack of critical 
parts. The group is a difficult one. To casual observation 
several of the species resemble one another almost to the degree 
of identity. Until they can all be studied as living plants there 
will be unsolved questions about them. At the present time, 
to my knowledge, the following species are in cultivation :— 
Rh. cuneatum, Rh. fastigiatum, Rh. favidum, Rh. hippophaeoides, 
Rh. idoneum, Rh. impeditum, Rh. intricatum, Rh. lapponicum, 
Rh. nivale, Rh. parvifolium, Rh. rupicolum, Rh. scintillans. 
Doubtless there are more, because more than one species has on 
occasion appeared in one seed pan. There is no little confusion 
amongst the cultivated forms. Rh. fastigiatum covers more 
than one species. Rh. Edgarianum as it has come to us includes 
two distinct plants, neither of them the right one, and therefore 
I have omitted it from this garden list. Rh. intricatum also 
is a name including two or three species of our gardens. Rh. 
nigropunctatum is not in the list. I have not seen it either alive 
or in herbaria, and doubt if it has ever been in cultivation. 
Rehder and Wilson say that the plant formerly grown under 
the name is Rh. intricatum. 
I have seen, thanks to the kind co-operation of M. Lecomte 
in Paris, types of aJl Franchet’s species excepting Rh. nigro- 
punctatum and Rh. ramosissimum ; I have also seen specimens 
of all Rehder and Wilson’s species, unfortunately only in small 
amount ; Maximowicz enriched our collections many years ago 
by samples of his species, and along with all these I have had the 
rich spoil of the exploration by Forrest and Kingdon Ward. If 
I attempt to sift the characters of the species and to find affini- 
ties within the series, it is mainly in the hope of helping culti- 
vators by giving them at least in some cases easily observed 
marks of distinction by which to recognise their plants. Before 
proceeding to this analysis it may be helpful if I elaborate the 
brief outline given above of the characters of the series. 
We have in the series carpet-forming species which are well 
illustrated ‘by the name-species of the series Rh. lapponicum 
and by the Himalayan Rh. nivale, plants which live for a 
greater part of the year under snow, and by Kh. tapetiforme and 
Rh. drumonium. Of the taller shrubby forms, Rh. parvifolium, 
Rh. capitatum, and others represent those which have elongated 
tapered leaves of some size. The dwarf cushion habit with 
gnarled twisted branches is a growth form that is adopted 
