Therefore, that from the latter a decided form, or a mixture of several forms, is to be found sown on one spot, is no proof of their generic connection with one which has been sown for the purpose of experiments; and the matter will only be more confused if we call imagination to our aid, and place the forms which are found near one another, according to a real or fancied resemblance, in a certain series of development. All those statements on the sphere of form and connection, which have for their basis such a superficial work, and are not based on the clear exposition of the continuity of development, as by the origin of the connection of the _Mucor_ with _Penicillium_, _Oidium lactis_ and _Mucor_, _Oidium_ and _Penicillium_, are rejected as unfounded. A source of error, which can also interfere in the last-named superficial method of cultivation for experiments, is, viz., that heterogeneous unwished-for spores intrude themselves from without, among the seed which is sown, but that has been until now quite disregarded. It is of great importance in practice, but in truth, for our present purpose, synonymous with what we have already written. Those learned in the science of this kind of culture lay great stress on its importance, and many apparatuses have been constructed, called "purely cultivating machines," for the purpose of destroying the spores which are contained in the substratum, and preventing the intrusion of those from without. The mixture in the seed which is sown has of course not been obviated. These machines may, perhaps, in every other respect, fulfil their purpose, but they cannot change the form of the question, and the most ingeniously constructed apparatus cannot replace the attention and intellect of the observer.[B] Two distinct kinds of phenomena have been grouped under the term "polymorphy." In one series two or more forms of fruit occur consecutively or simultaneously on the same individual, and in the other two or more forms appear on a different mycelium, on a different part of the same plant, or on a matrix wholly distinct and different; in the latter case the connection being attested or suspected circumstantially, in the former proved by the method suggested by De Bary. It will at once be conceded that in cases where actual growth and development substantiate the facts the polymorphy is undoubted, whilst in the other series it can at best be little more than suspected. We will endeavour to illustrate both these series by examples.