ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 
191 
genera of restricted range, we see the evidence of their geographical dis- 
tribution strongly confirming the opinion of students of their structure, 
that they are the most primitive of the dragon-fly sub-families.” 
How Flowers attract Insects.* — Another contribution by Prof. F. 
Plateau to our knowledge of this subject treats of anemophilous flowers, 
and of such as have but little natural colouring ; and he draws from his 
observations conclusions favourable to his view already expressed, that 
insects are but little attracted to flowers by the sense of sight, chiefly by 
that of smell. 
In all the species of strictly anemophilous flowers observed, 17 in 
number, all of some green or brown tint, and not visited at all by insects 
in nature, it was sufficient to place on them fragrant artificial nectar in 
order to attract numbers of insects of different kinds, chiefly Diptera and 
Hymenoptera. The author enumerates moreover 91 species, belonging 
to a great variety of natural orders, which are entomophilous, but in 
w r hich the flowers are inconspicuous, either bright green, very pale green, 
or brown. 
In summing up his conclusions in the concluding part of his paper, 
Prof. Plateau recapitulates his reasons for regarding the sense of sight 
as only a secondary one in attracting insects to flowers. If the con- 
spicuous parts of flowers are removed insects still continue to visit them, 
and they are at once attracted to flowers which they do not ordinarily 
visit if these are smeared with honey. Where a species varies in the 
colour of its flowers, they exhibit neither preference nor antipathy to one 
colour over another. Artificial flowers made of paper or calico, and very 
closely resembling real flowers, do not attract insects, while they do if 
made of leaves which have a vegetable scent. Inconspicuous flowers 
completely hidden by foliage are still abundantly visited by insects if 
scented. 
Mimetic Attraction.! — Dr. F. A. Dixey has shown that the process 
of mimetic assimilation may start from a given form and proceed along 
several divergent paths. Thus from Pieris phaloe five divergent series of 
mimetic modifications could be traced, each passing through a graduated 
series of closely allied forms until it terminated in a Pieris or Mylothrix , 
bearing a more or less intimate relation with some protected form or 
forms of entirely different affinities. In these instances the model to- 
wards which the series tended was a member of a “ Mullerian group ” — 
an association of inedible species sharing a common notoriety of colour- 
ing. The ordinary Batesian mimicry between edible and inedible can 
exist only when the numbers of the mimic are insignificant compared 
with those of the model, whilst a Mullerian group gains in strength with 
every fresh accession. The attractive power in Batesian mimicry acts 
only from model towards mimic, whereas in Mullerian association the 
attraction is mutual and tends to produce reciprocal changes. This 
reciprocal mimetic attraction is probably the more important of the two. 
As a consequence of the keen struggle in the Neotropical fauna, scarcely 
any conspicuous form is completely isolated. If edible, it is generally 
* Bull. Acad. R. Sci. Belgique, xxxiv. (1897) pp. 601-44, 847-80. Cf. this 
Journal, 1897, p. 121. f Trans. Ent. Soc. London, 1897, pp. 317-31 (1 pi.). 
