TRANSACTIONS OF SECTION D. 1083. 
separate from those within this edge, and from those in the centre of the pod. 
(The shape of the seeds alters from the margin to the centre, but the plants 
resulting from those on the margin and the centre, if impregnated alike, do not 
vary). 
Thus in composite flowers every individual flower can be made to produce 
distinct variations on the same flower-head. 
An experiment was tried as to whether it required more than one pollen grain 
to impregnate a flower. A white-flowered dahlia was selected as the seed-bearer, 
and this was crossed with pollen from a deep red flower. When simply crossed 
this produced plants having blooms from pink to red, and three plants with white 
flowers, whilst if crossed with pollen containing four times as many grains from a 
white flower as from a red one, the result was in 172 seedlings, 79 were pure 
white, and all the others pale in colour, none being nearly as dark as the red 
flower from which the pollen had been taken. 
The plan adopted to obtain certain proportions of pollen grains was after con- 
sideration as follow :— 
Four brushes were filled with pollen from a white dahlia, and one brush from 
a red dahlia, and this was collected from these five brushes on to a larger brush, 
and this larger brush was the one used to impregnate the example selected. 
It is easy to get an approximate idea of the colours to be obtained by certain 
crosses by the mixture of moist paints or dry powders, and in the case above 
mentioned, the mixture of four times as much white as red was singularly borne: 
out, as no seedling produced flowers darker than this mixture. Those who are 
practically aware of the difficulty of raising white seedlings even from the seeds of 
a white dahlia will understand the success of these experiments, and will be con- 
vinced that a number of pollen grains are necessary in order to etfect impregnation, 
or at all events have the power of acting together, 
4. On the Occurrence of Fungi in the Roots of Orchids. By J. MacmILLan. 
It has long been known that the mycelia of fungi occur in the roots of orchids. 
Prillieux! saw them in Neottia Nidus avis, Reinke. notices their occurrence in 
Corallorhiza and Epipogon, and several other writers have noticed them. De Bary, 
in a lecture on symbiosis given at Innsbruck in 1879, also mentions the fact. 
As yet no systematic study of the phenomenon has been made. The difficulties 
attending such study are great. I propose here to record the results of my 
examination of several species of orchids, chiefly epiphytal orchids. 
Epiphytes. 
Cattleya Gaskelliana. Plant in full flower, few aerial roots, no trace of mycelia 
in aerial roots one year old, no trace in fresh shoots of this year. 
C. Mendelit. Plant not in flower, in one aerial root mycelia found. 
C. Mossie. Plant examined three months after flowering, many aerial roots 
containing mycelia. One root coming from a pseudobulb pushing out a flower 
contained abundant mycelia. 
Oncidium Forbesi. Plant in full flower; no trace in any part of any aerial 
root. 
O. Forbesi. Another plant not in flower; no trace in any aerial root. 
O. crispum. Plant not in flower, no trace of mycelia. 
O. Kramert. Plant not in flower, no trace of mycelia. 
O. fuscatum. Plant not in flower, no trace of mycelia, 
Stanhopea grandiflora. Plant in full flower, perfume of flower very powerful, 
aerial roots abundant and mycelia in these abundant. 
S. oculata, Plant examined one year after flowering, many withered aerial 
roots in which no trace, many new aerial roots with mycelia abundant. 
Odontoglossum trivmphans. Plant not in flower, numerous pseudobulbs and 
aerial roots springing from them. No trace of mycelia. 
1 Annales des Sciences Naturelles, 1856. 2 Flora, p. 145, 1873. 
