THE TirUEE FORMS OF I.YTUErM SALTCAI!rA. 195 



fact is iu itself curious, and shows by wliat insensibly graduated 

 steps nature moves. If this tendency were carried out the 

 mid-styled form would becomo a female, depending for its fer- 

 tilization on two sets of stamens in the long- and sliort-styled 

 forms ; and these two forms would reci[)rocally fertilize each other 

 like the two forms of Priiinda or Linum ; but tlicre would be no 

 approach to a dioecious condition. 



As the case of the trimorphie species of Lythrum is so com- 

 plicated, and as it is easier to perceiYO the relations of the sexes 

 in the animal than in the vegetable kingdom, it may be worth 

 while to give, before concluding, a somewhat elaborate simile. 

 We may take the case of a species of Ant, and suppose all the 

 individuals invariably to live in three kinds of communities ; in 

 the first, a large-sized female (not to specify other difterenccs) 

 living with six middle-sized and six small-sized males ; in the 

 second, a middle-sized female witli six large- and six small-sized 

 males ; and in the third community, a small-sized female with six 

 large- and six middle-sized males. Each one of these three 

 females, though enabled to unite with any male, would be nearly 

 sterile with her own two seta of males, and likewise with two 

 other sets of males living in the other two communities ; for she 

 would be fully fertile only when paired with a male of her own 

 size. Hence the thirty-six males, distributed by- half-dozens iu 

 the three communities, woiild be divided into three sets of a 

 dozen each ; and these sets, as well as the three females, would 

 differ from each other sexually in exactly the same manner as 

 distinct species of the same genus. Moreover the two sets of 

 males living in the community of the extraordinarily fertile 

 middle-sized female would be less potent sexually than the males 

 of corresponding size in the two other communities. Lastly, we 

 should find that from the eggs laid by each of the three females, 

 all three sorts of females and all three sorts of males were 

 habitually reared^X)roving to demonstration that all belonged to 

 one and the same species. 



To appreciate fully this remarkable case of the reciprocally tri- 

 morphie species of Lytlirmn, wc may take a glance at the two great 

 kingdoms of nature and search for anything analogous. With 

 animals we have the most astonishing diversity of structure iu 

 the so-called cases of alternate generation, but as such animals 

 have not arrived at maturity, they are not properly comparable 

 with the forma of Lythrum. With mature animals we have 

 extreme differences Ln structure iu the two sexes; we have in 



