The same author alludes to certain changes in the same flower where two open carpels "were soldered together laterally, as was clear by the rudiments of two styles, the placenta being produced only at the two united edges, the outer margins remaining in the normal condition. This may possibly tend to the explanation of some cases of anomalous placentation, for the only indication of the true nature of the placentation is afforded by the two rudimentary styles, in the absence of which the spongy receptacle of the seeds must have been supposed to spring from the medial nerve." In other cases the placentas were parietal above, but axile at the base of the capsule, a striking instance of the facility with which axile placentation becomes parietal, the change being here effected by the prolongation of the axis, and the formation on it of a second whorl of carpellary leaves. In double flowers of _Primulaceae_ similar alterations in the placentation may often be observed. I have seen in _Primula sinensis_ sutural, parietal, axile, and free central placentation all on the same plant; nay, even in the same capsule the ovules may be attached in various ways, and transitions from one form of placentation to another are not infrequent. The late Professor E. Forbes describes[103] an instance of true foliar and true axile placentation in the same flower in _Vinca minor_. These and many similar changes, which it is not necessary further to allude to, are not so much to be wondered at when it is borne in mind how slight an alteration suffices to produce a change in the mode of placentation, and how frequent is the production of adventitious buds or of foliar outgrowths, as may be seen in the sections relating to those subjects and to Substitutions. It will be remembered, also, how, in certain natural orders, under ordinary circumstances, considerable diversity in placentation exists, according as the margins of the carpels are merely valvate or are infolded so as to reach the centre. Often this diversity is due merely to the changes that take place during growth; thus, the placentation of _Caryophylleae_, _Cucurbitaceae_, _Papaveraceae_, and many other orders, varies according to the age of the carpel, and if any stasis or arrest of development occurs the placentation becomes altered accordingly.