154 DIVISION II—COURSE OF DEVELOPMENT OF FUNGI, 
as monosporous sporangia completely filled by the spore, and most comparable with the 
sporangiola of the Thamnidieae. These interpretations, which are not in accordance 
with the facts, have evidently arisen from a perception of the truth, that all the gonidial 
formations in question are homologous, coupled with the erroneous notion that 
homologous spores must necessarily originate by exactly the same mode of cell- 
formation. We naturally gain nothing more from such interpretations in this case 
than in others of older date which were noticed above on page 116. The homologies 
are brought out quite simply in our case, as has been shown above, without these 
artificial helps; and conversely the case before us shows with special distinctness that 
members and cells may be homologous, though all the facts observed with regard to 
them do not belong exactly to one and the same category of the scheme of cell- 
formation which is accepted at the time. 
Section XLIV. Aceessory gonidia. In a number of species, in Sporo- 
dinia grandis for example, Brefeld’s Mucor Mucedo, Rhizopus nigricans and 
Chaetocladium, the cycle of forms is exhausted by the appearance of the members 
which have now been described; but in other species gonidial formations occur 
in addition to the typical ones, some of which by their form and structure are 
eminently characteristic of species and genera. It would seem that they are always 
found on starved or old mycelia, but the conditions of their formation have not been 
in every case clearly ascertained. It would be difficult to find any other general 
name for them than the one here chosen. Many of them have been called by Van 
Tieghem chlamydospores and stylospores, others gemmae, &c. It will be most 
convenient to choose a suitable name for each case as it presents itself. 
Characteristic accessory gonidia are the acrogonidia of Mortierella and some 
species of Syncephalis. They are solitary and acrogenously abjointed on the 
summit of slender cylindrical branchlets of the mycelium, and are spherical in 
shape and usually of considerable size in Mortierella (as much as 20 » or more 
in diameter), but small in Syncephalis (6 »); they have a thick episporium with 
its surface marked with warts or spikes in a manner characteristic of the different 
species, and they emit germ-tubes which may develope into the normal mycelium. 
They are formed in Mortierella on erect branches of the mycelium which are either 
isolated or are united into small umbels; in the species of Syncephalis from short 
stalks which are arranged in dense racemes springing at right angles from a fusiform 
swollen portion of a mycelial filament. A very remarkable instance of the formation 
of accessory gonidia is that described by Cunningham in his genus Choanephora, 
in which the heads of basidia mentioned above must, from the mode of their 
occurrence on the normally developed mycelium, be certainly regarded as the typical 
gonidiophores, though the germination of the zygospores has not yet been observed, 
and cannot therefore confirm this view. But erect simple sporophores make their 
appearance on old and starved mycelia by the side of meagre heads of basidia, and 
form at their summit a spherical mucor-sporangium having a warted outer wall, a 
slightly convex basal wall, and containing a number of ellipsoid smooth-walled 
spores, which can germinate and produce a normal mycelium. 
Another form of accessory gonidia is known under the name of gemmae or 
brood-cells. Their ordinary mode of production is that short pieces of a mycelial 
tube or gonidiophore, which is rich in protoplasm, become delimited by transverse 
walls to form cylindrical or nearly spherical or ovoid or pear-shaped or similarly 
