portions of the stamen. It may be seen
in some forms of double columbine,[310] in which the connective forms a
tubular petal or nectary, and in double petunias and fuchsias. When it
occurs, the true anther-lobes are usually atrophied, and little or no
pollen is formed.

An occurrence of this nature in _Tacsonia pinnatistipula_, in
conjunction with the partial detachment of the stamens from the
gynophore, led Karsten to establish a genus which he called
_Poggendorffia_.[311]

From the subjoined list of genera in which petalody of the stamens, in
some form or other, has been observed, it will be seen that it happens
more often in plants with numerous distinct organs (Polypetalae,
Polyandria, Polygynia, &c.) than in other plants with a smaller number
of parts, and which are more or less adherent one to the other. The
tendency to petalification is, moreover, greater among those plants
which have their floral elements arranged in spiral series, than among
those where the verticillate arrangement exists; and in any given
flower, if the stamens are spirally arranged while the carpels are
grouped in whorls, the former will be more liable to petalody than the
latter, and _vice versa_. It has been before remarked, that this
condition is far more common in plants whose petals, &c., have straight
veins, like those in the sheath of a leaf, than in those the venation of
which is reticulate, as in the blade of the leaf. It must also be
remembered that in the same genus, even in the same species, different
kinds of doubling occur. Familiar illustrations of this are afforded in
the case of anemones, columbines, fuchsias, and other plants.

The existence of "compound stamens" in some flowers, as pointed out by
Payer, and others, and the researches of Dr. Alexander Dickson, confer
additional importance on the subject of petalody, and necessitate the
examination of double flowers with special reference to these compound
stamens, and to the order of their development.[312] The presence of
these compound stamens affords a satisfactory explanation of the
appearance in some double _Malvaceae_, wherein the tufts of adventitious
petals are very liable to be mistaken for buds, produced by axillary
prolification in the axils of the petals, but which are in reality
compound and petaloid stamens. At other times, however, true axillary
prolification exists in these flowers; but then the supplemental florets
have always a calyx, which is wanting in