( xc! ) 



formulated by l)iologists, the case of Pack>/lia tlnreeta becomes of general 

 importance, counselliug caution, since darci'ta has the right harpe more reduced, 

 and hence ])roves either that there is no inherited tendency in tlie Sesiinae to 

 a left-sided reduction, or, if there is, tli.at the opposite development has taken 

 place in spite of that tendency. It may be a matter of little moment to an 

 outsider that the only species in the second divisimi of the Sphingidae which 

 has in the ]inpa a projecting toiigne-case, as it is found in many Aclierontiinae, 

 is RInimholaba actens ; but the matter becomes ditfereut if we explain that, 

 firstly, tlie jiresence of a free tongue-case in Klii/iicholubu and Acherontihme is 

 not a sign of relationsliip, as tlie organisation of tlie insects proves, and hence 

 demonstrates the jiossibility of the ap])earaiice of a similar consjiicuous character 

 in very distantly related gronjis, and, secondly, that the similarity again cannot 

 be mimetic, since the pupae arc buried. 



In chemistry and physics minute research discloses the secrets of Natur^'. 

 In biological sciences minuteness of research is as much required as there, if 

 we wish to understand the phenomena of life. Systematic work can do much 

 towards that end ; it is the only means of checking off, proving and disproving, 

 generalisations. But to serve as a reliable censor of conclusions, the observa- 

 tions must be e.Kact, and therefore minute, since exactness cannot be attained 

 without niiiiutencss of inijuiry. A siin[)liticatioii of systematic research by 

 uarrowing it ilown, as advocated in certain quarters, to the solution of the 

 questions " obvious difference or no obvious difference, hence species or no 

 species, genus or no genus," takes the life out of it. But if these questions 

 are made subordinate, and therefore systematic work subservient, to the higher 

 issues of biology, the dry detail of distinguishing characters and the apparently 

 futile labours of the describer of species and other classificatory units become 

 all-im[)ortaut in the science of life, as supplying sound criteria, where otherwise 

 a lively imagination might run wild and substitute plausible a>snmptions for 

 facts. Looking from this jioiiit of view at the diagnoses of the genera and 

 species of Spliitiqidac it will be fnund that they are a source for information of 

 a general cliai'acter. 



If we compare the jiedigrees (facing i)p. 24, 4(19, etc.), with one aiiotlier, and the 

 characters there mentioned with the fuller descriptions given in the body of 

 the Revision, the first thing which strikes one is the frequent repetition aud 

 reappearance of the same diagnostic character (for instance, the absence of a 

 pidvillus, or the sjiiuosity of the tibiae). Whether our classification is accepted 

 as valid or not, the fact remains that there is very frequently a close .agreement 

 in one or more characters between sjiecies and genera of near or of distant 

 relationship in the Hawk Moths, an agreement which renders the classification 

 of the Spldngidae especially difficult. Since the agreement refers to many details 

 iif many organs, as well as to the colour and structure of the scaling aud the 

 sha|je of tlie biidy and wings — which give the insect its habitus — the Spliiiii/idae 

 tilrlli^h iiidci'd iui aliiindaiice of ni;ilerial lor a Study of Similarity, towards 

 which we oll'ci- ii lew reniiirhs. 



