FAMILY MARMORACEAE 1209 



(Note: The genus Citrivir (first named species, Cilrivir psorosis Fawcett, Phyto- 

 path., SI, 1941, 357) was proposed by its author as a genus pro tempore with the avowed 

 purpose of accommodating viruses causing diseases in species of the plant-host genus 

 Citrus. It appears to have been implied by the term genus pro tempore that evidences 

 of natural relationship, when discovered, would permit even the first-named species 

 of this genus to be assigned elsewhere. On the assumption that a permanent genus is 

 nothing more than a type species and such other species as may be added to it by one 

 or another author, it must be felt that a genus pro tempore, however convenient as an 

 expedient, cannot become a permanent genus under any circumstances, because its 

 first -named species would appear not to be a permanent part of the genus and so 

 intended not to be a true type-species. Without a type species there would seem to 

 be no permanent genus concept . 



The system by which the term Citrivir was coined (explained by its author as use of 

 the genitive of the host-genus name, Citris, plus vir, signifying virus) seems in itself 

 acceptable, for it is commonly agreed that a generic name may be made in an arbitrary 

 manner. It may be noted that use of the stem of the host -genus name (Citr-) with 

 connecting vowel i and suffix -vir, possibly a more orthodox procedure, would have 

 given the same result in the present instance. The original definition of the term 

 Citrivir might be thought to be repugnant as disregarding concepts of natural inter- 

 specific relationships that are essential to the spirit of binomial nomenclature. Were 

 the genus to be regarded as permanent rather than pro tempore, however, the scope of 

 the genus would come to be wholly changed by usage, when, with passage of time 

 related species would be added to what in this case would be a type species, without 

 regard to the unorthodox intent of the original definition but .solely in accordance with 

 similarities between viruses. A generic concept need never be accepted as rigidly 

 defined, whether initially, as has been attempted in this case, or upon further experi- 

 ence, because a genus may still grow by the addition of closely allied new species 

 beyond any limit that may be set. On this account an original, or any subsequent, 

 definition may be regarded as subject to unlimited change so long as the type species 

 is logically retained. The form and definition of the term Citrivir would not, there- 

 fore, militate against its continued use. Its avowedly temporarj'' status alone seems 

 decisively to do so. 



The originally monotypic genus Rimocortins , publishetl in the following year, was 

 defined only by the combined generic and specific description, and was not referred to a 

 family by its authors. The type, because at first the only species, Rimocortius kwan- 

 zani, is the flowering-cherry rough-bark virus. This type species might well be asso- 

 ciated with the species Citrivir psorosis, citrus-psorosis virus, discussed above, 

 both affecting bark principally, though foliage also to some extent. Although the 

 genus Citrivir was named in 1941 and Rimocortius not until 1942, the first was intended 

 as a temporary assemblage only, as above indicated. It would seem appropriate, there- 

 fore, to include the virus that was known temporarily as Citrivir psorosis in the per- 

 manent genus Rimocortius Milbrath and Zeller and to assign this genus to the family 

 Marinoraceae.) 



Key to the species of the genus Rimocortius. 



I. Affecting cherry. 



1. Ri7nocortius kwanzani. 

 II. Affecting Citrus. 



2. Rimocortius psorosis. 

 III. Affecting pear. 



3. Rimocortius pyri. 



