676 A MANUAL OF THE PENICILLIA 



shapes and measurements of those in typical Penicillia to long subulate 

 tubes. In some species, as also in the polyverticillate, a series of reflexed 

 or "rhizoid"-like branches often forms a claw-like support at the base of 

 the conidiophore. 



From the standpoint of relationship with PenicilUum, the process of 

 conidium formation becomes significant. The typical cell of the group 

 is the sterigmata which cuts off conidia from its apical tube. Microscopic 

 examination shows that the conidia are cut off successively from the tips of 

 sterigmata just as they are in PenicilUum, but typically do not adhere in 

 chains. Instead they slip back and become enveloped in slime. In study- 

 ing various areas and different ages of petri dish colonies of certain strains, 

 transitional steps are found. In some areas the conidia remain for a tim6 

 in chains, in other areas of the same colony penicilli show the Clonostachys- 

 type of conidial columns in which the conidia are arranged diagonally and 

 in still other areas the conidia are found massed in the typical slime balls. 



Correlation of these observations with cultural conditions shows the 

 Penicillium-like arrangement of conidia to be primitive and regularly 

 present in conidial structures during the process of conidium formation. 

 One species, Gliodadium vermoeseni (Biourge) Thom. produces separate 

 chains of conidia that retain their identity as in PenicilUum. In another 

 species, G. catenulaium Oilman and Abbott (1927), the conidia may remain 

 end to end in chains, but the chains become adherent into enslimed columns 

 bound more firmly than in any of the true PeniciUia (fig. 170B). In G. 

 deUquescens the conidial apparatus is regularly penicillate, but the conidia 

 invariably accumulate into conspicuous balls of abundant slime (figs. 170C 

 and D). The extent of the deliquescence thus determines whether conidia 

 in several chains will remain in position as produced, or will adhere into wet 

 columns as figured by Oilman and Abbott for Gliodadium catemdaium, or 

 whether every ccnidium will slip back slightly toward the base of the chain 

 to assume a position diagonal to the axis of the column, as in Clonostachys, 

 or whether no cell-to-cell relationship will be maintained and all conidia will 

 collect into a composite globule of slime in the manner regarded as typical 

 of the genus Gliocladium. 



Three more or less contrasting series of Gliocladium are commonly en- 

 countered in culture, namely: 



(1) Gliocladium roseuni series. This rosy or salmon series shows a 

 fairly complete gradation from forms near G. vermoeseni (Biourge) Thom 

 with conidial structures hke a true PenicilUum, through types simulating 

 Corda's Clonostachys araucaria with the end-to-end relations of conidia 

 broken and each cell slipped half its length backward, to forms like 

 G. roseum in which the slime balls are usually well established. 



