290 
Transactions of the Royal Society of South Africa. 
closely associated in habitat with the suggested parent forms. The occur- 
rence of hybrid A was noted some four years ago on the farm Kleinberg in 
the Tulbagh District. 
Since then I have collected specimens each year on this and neighbouring 
farms, and have noticed them at the Tulbagh flower shows as exhibits from 
various parts of the district. Apparently this form has been found more 
abundantly than is the case with the hybrids mentioned above. 
I have made a detailed study of the form from about a dozen spikes 
during the present flowering season, and have had the special advantage of 
working on fresh material. 
Specimens of hybrid B described in this paper were noted for the first 
time this year on the farm Kleinberg and also as exhibits at the Tulbagh 
flower show. The supposed parent forms were associated in abundance 
with the hybrid. 
The hybrid nature was suggested by the intermediate nature of the size 
and colouring of the flower and the form of the stem sheaths. This 
appears to be borne out by the results of more detailed study of the floral 
and vegetative parts. 
It will be noted from an examination of the subjoined comparative 
tables that in each case the hybrid is an intermediate or blend between the 
two suggested parent forms, both as regards vegetatiye and floral structures. 
An attempt was made to obtain average measurements of the entire plant, 
spike, leaves, etc., of the parent and hybrid forms, but there is so wide 
a variation in the individuals of all the types that it would be misleading 
to quote any average figures. 
Further details for the comparison of the hybrids are found in the 
diagrams. 
The following observations on the parent forms in each case in 
conjunction with the natural association of the same, support the 
probability of the hybrid origin of the two types described : 
(1) Frequency of pollination, (I have noted in numerous instances 
the removal of one or both pollinia and the presence of pollen on the 
stigmatic surfaces of the parent forms, particularly in the case of 
S. erectum.) 
(2) The sweet scent of the flowers in all the parent forms concerned 
and especially so in the case of the inconspicuous flowers of S. bicorne. 
(3) Presence of free honey in the spurs of all the parent forms. 
(4) The long life of individual flowers. 
(5) Insect pollination. (I have several times noticed insects visiting 
the flowers and in some cases observed pollinia attached to these insects. 
That this is not altogether occasional is suggested by the abundance of 
spiders among flowers and bracts, and some of these show a striking 
adaptation in colouration to that of the flowers in S. erectuyn.) 
ll 
