16 PARKTN : FUNGI PARASTTIC 



no diagnosis of now ones has been undertaken, as this on the 

 one hand is more the work of a specialist, and, on the other 

 hand, their in vestigation is not sufficiently advanced to permit 

 of this. A hasty formation of new species,specially of f ungi, 

 is to be deprecated. It seems preferable to err rather in the 

 opposite direction, grouping together forms presenting slight 

 differences, till further study warrants them being split into 

 distinct species. Small deviations in size, shape, or colour, 

 unless definitely shown to be constant, do not appear to be 

 sufficient for the separation of species of taxonomic value. 

 Colour as a systematic character should be used with caution, 

 because it can alter with age, and has often faded in material, 

 the examination of which has been postponed for some time 

 after its collection. Then again no two observers are likely 

 to describe the same tint in exactly the same words. So in 

 order to arrive at uniformity in systematic colour descrip- 

 tions a standard chromatic scale should be used as re- 

 commended by Guegnen.* 



The recognition of what are called "biological species" 

 amongst parasitic fungi renders still more difficult the 

 separation of what should rank as taxonomic ones. To take 

 a hypothetical case, the Microcera on Aonidia differs in 

 certain external features from that on Aspidiotus ; this is 

 perhaps accountable for through a difference in the general 

 structure of these coccids. If, however, the Microcera on 

 Aonidia should be shown to be incapable of directly affecting 

 Aspidiotus, would it be then justifiable to separate the two 

 forms of this fungus as distinct taxonomic species ? 



All these scale-fungi are either complete Ascomycetous 

 formsorincomplete conidialones. The former are described 

 first. The latter are provisionally placed in the Fungi-im- 

 perfecti, an artificial group the individuals of which are 

 generally considered as conidial stages of various Ascomy- 

 cetes. But as their ascus-fructifications arc as yet unknot a 

 they cannot with any degree of certainu l>e elassilied with 

 them. 



Q-odgnen. loo. <■/'/.. p. 9, 



