PRINCIPLES, CANONS, AND RECOMMENDATIONS. 55 



associations which are distributed [only] to members or not at all; and in 

 brief lists, catalogues [especially sale catalogues], or pamphlets indepen- 

 dently issued, insufficiently distributed, or not to be found on sale." (Rep., 



p. 46.) 



The question of the restriction of the nature of the channels of publication 

 through which new species and genera, and changes in nomenclature, should 

 be made public, is considered by Mr. Ball, and was even included among the 

 subjects covered by his circular, the replies to which were to the effect that, 

 while such restriction would be very desirable, it seemed impracticable; an 

 opinion reluctantly concurred in by Mr. Ball himself. 



"It is clearly," Mr. Ball continues, "the duty of every publishing author 

 to concur as far as possible in the suppression of methods leading to confu- 

 sion," and to comply with recommendations " intended to lead toward this 

 result." 



CANON XLVIII. The reading of a paper before a scientific 

 society or a public assembly does not constitute publication, 

 and new genera and species first announced in this way date 

 only from the time of their subsequent and irrevocable pub- 

 lication. 



REMARKS. It often happens that papers are read before a scientific body 

 which are never printed. No one would claim publication in such cases. 

 Often many months elapse between the reading of a paper before a society 

 and its publication in the proceedings of the society. Credit for original dis- 

 covery may be thus secured; but, in deference to the fundamental principle 

 of fixity in nomenclature, new names or changes in nomenclature proposed 

 in such papers obviously cannot be allowed to antedate actual publication. 



CANON XLIX. The date borne by a publication is presumed 

 to be correct till proved otherwise ; although it is well known 

 that in many instances, as in the proceedings or transactions of 

 societies, and in works issued in parts, the date given is not 

 that of actual publication ; and when this fact can be substan- 

 tiated, the actual date of publication, if it can be ascertained, 

 is to be taken. 



REMARKS. It is notorious that the dates on the title-page of the com- 

 pleted volume of works issued in parts often antedate sometimes postdate 

 the actual publication of the different parts, or are otherwise erroneous. 

 Also, that the volumes of proceedings of learned societies not unfrequently 

 bear simply the date of the period or year to which they relate, even when 

 rot published till months, and sometimes years, after the ostensible date ; 



