• 



INTRODUCTION. 



VII 



of the true stigma, but has often little or none of the viscid parenchyma c tanuteristi- 

 of that organ. These peculiarities in the nature of the stigma and the shortness of the 

 style are apparent in the gall flowers of many specie ^ from a very early stage. Th< y 

 are not consequences of the deposit of the egg of an insect in the ovary, but, as Count 



Solms-Laubach points out (Bot. Zeituny, I.e.), such original peculiarities in the stylo and 

 stigma of the gall flower may rather determine the selection of it by the insect as the 

 nidus for its eg<^. There are, however, many species of Ficus (more - specially in 

 the group Urwtiyma) in which the gall and fertile female flowers are not charaet rised 1 

 any marked differences in the form of tyle and stigma, and it is only by cutting the 

 ovaries open that the two can be distil igui lied. 



Now there is probably nothing in itself very remarkable in the mei \ occ ur r e nce in 

 the genus of numerous flowers having the general form of females, which yet, b\ reason 

 of certain peculiarities in their structure, are incapable of fertilisation by pollen and are 

 practically barren, while at the t one timo their very structural defects iit them foi 

 becoming the nidus for the eggs of special insects. Hut when the manner in which 

 these malformed female flowers are disposed in the recept:«los is inquired into, it 

 becomes clear that, through the interposition of in ects, these malformed females may pi 

 a most important part in the life-history of many species of the enus. En all the s]>ecios, 

 except those included in the section Urmtignia, the gall flowers occupy the sumo recc] 

 tacles as the males, while the fertile female How -? occupy dill nt leceptacfc In 



other words, the majority of the peci< 8 have two di tinct sets of i pta» Irs— one t 

 containing male and gall flowers, but no fertilo female flowers; and ai >ther et contain- 



in «■ only fertile female flowers without any trace of either male or gall fl<>w< >. The 



proportion of males to gall flowers in r oeptacles of the former kind varii -. In all 

 (excepting the Urostujmus just mentioned) it is the rule to find the mates confined to a 



zone of greater or less width at the apex of the receptacle just under the scales which 



close its mouth. Sometimes this zone is very narrow indeed, and < insists of only a 



single row of male flowers, and that row not always a com pi- to one; the remaining part 



of the interior of the receptacle being occupied by gall flowers. In by far the majority 



of cases these two kinds of receptacles, so physiologically distinct, are and i ng ni sh able 



by external characters, and they are both borne by the same individual plant. Ti J 

 look exactly alike until one cuts them open and examines their content . The m< 

 notorious of the few exceptions to this rule is the common eatable tig {Ficus Carina), in 

 which species the male and gall flowers occupy elongated receptacles borne in one set of 

 individual trees, while the fertile female flowers oc upy more or less globular rec ptach 

 which are borne by a different set of trees. So differ nt in appearance are the tw« 

 kinds of receptacles in F. Carica, that the trees bearing them (although they have similar 

 leaves) have almost from time immemorial been considered distinct specie , known by 

 distinct names-the former being called the Capnfig, the latter the Fig. A vague idea of 



