246 



SCIENCE. 



[Vol. XX. No. 50^ 



I t^hall try to consider the botanic trinomial, not from the 

 ethical point of view as Mr, Townsend seems to have done, but 

 from the taxonomic strictly. 



We find it convenient to give a name to a plant simply because 

 the use of the name serves to call up an aggregate of character- 

 istics vhen we wish, without the necessity of detailing those 

 characteristics. The whole matter is one of convenience simply, 

 and a name means nothing more than this. 



It has been pretty universally agreed that it is more convenient 

 to have a binomial name than a monomial one, for by this means 

 we are enabled easily to group our plants, the first name serving 

 to call to mind the aggregate of characteristics of the group 

 (genus), possessed often by many sub-groups (species), and the 

 second those characteristics possessed to a greater or less f xtent 

 by the individuals that go to make up the subgroup. 



So far this seems to be reasonable enough, and, followint; the 

 same lines, should we choose to add a third name to our binomial, 

 making it a trinomial, we should naturally do so for the purpose 

 of segregating these sub-groups into still smaller ones (varieties). 

 On this line the addition of terms might rationally be continued 

 to the extent that the facts of observation would warrant. . 



But we find in the cle facto botanic trinomial a mixture of two 

 taxonomic principles, instead of the rational following out of the 

 single line indicated by adding to the monomial the second term. 

 Usually the third term is added as a compromise with existing 

 fact, simply to avoid the possibility of having two homonomic 

 binomials, and consists of the name of the person who first pub- 

 lished the binomial. It is evident that this addition of such a 

 third term sei'ves a purpose only in comparatively rare cases ; in 

 the vast majority, were it not for the fear that some future comer 

 would see fit to use the same binomial to designate another 

 plant, it would be, as a name, useless. But at present the addi- 

 tion of the author's name is essentially a part of the naming of 

 the plant. 



It is this third name, and comparatively useless one, that is the 

 cause of much of the trouble of the botanic taxonomists. Many 

 seem to feel that this serving as a compromising tailpiece, the 

 necessity for which it is confessedly the aim of the botanic world 

 to do away with altogether, is an honor. And for this reason 

 there is strife in a large class of cases as to the third name to be 

 added to the binomial. For consider the following specific case. 

 Hooker and Arnott noiice a plant, which, in their judguient, is a 

 member of the large group of plants that has been called Malva. 

 They therefore give it the binomial name Malva malachroides, 

 and first publish the characteristics which that name is to call up. 

 Afterward Gray considers that the plant cannot belong to ths 

 group called Malva, and so gives the same plant the name Sidal- 

 eea malachroides. More recently Greene finds that the plant can 

 be neither a Malva nor a Sidalcea, and calls it Hesperalcea mala- 

 chroides. 



Now suppose we have an individual of this group and wish to 

 give it the most convenient name. For the name of the main 

 group undeniably it matters not which of the three names we 

 choose; if we have had the opportunity of studying the plant 

 carefully our choice will be determined by the observed facts and 

 our own judgment. Personally, in the present case, I chose to 

 call the plant Hesperalcea. For the second name there is no 

 choice, the three authors having given it the same. (Elad there 

 been a diversity of names here, the name first given the plant 

 would have been chosen, not because this is " just," or " right," 

 but because by this artificial rule we obtain a permanent factor in 

 the name, without fossilizing individual opinion at all regarding 

 the affinities of the plant.) 



We now come to the third name, only added, remember, from 

 the fear that some one has called or will call some different plant 

 Hesperalcea malachroides. Here custom is divided, and many 

 would write H. malachroides, S. and A., and others if. mala- 

 chroides, Greene. It is for us now to determine which of these 

 names is the most convenient.' The person to whom we wish to 



1 I liave not considered the writing of H. malachroides (H. and A.) Greene, 

 as the parenthetical term is no more an essential part of the name than the 

 date of publication or twenty other particulars which might occur in a mono- 

 graph on the plant. 



communicate the idea, H. malachroides, upon seeing tbe tri- 

 nomial H. malachroides, H. and A., naturally turns to the works; 

 of H. and A. to find the summing up of the characters of the- 

 plant. But here he is met with an insurmountable difficulty. 

 He can find no trace of it. Let him look for malachroides, per- 

 chance Mr. Townsend would say. But it is easily possible that 

 H. anil A. have described five species by the name of malachroi- 

 des. On the other hand, suppose we write H. malachroides, 

 Greene, the person wishing to know of this plant would turn \o 

 the works of Greene and there would find the reference to Malva- 

 mala^hroides, H. and A. , which would enable him to find the- 

 original description of the plant and thus obtain the idea which 

 we wished to convey. 



It seems plain enough then that the third name of this trinomiar 

 from the standpoint of convenience should be Greene and not H. 

 and A.. 



Mr. Townsend disposes of this difficulty in tbe following- 

 words: — 



''I would write Metsgeria pubescens sohrank, . . . and make 

 no more ado or trouble about it. . . . This signiSes always that 

 the authority named described the species originally and originally 

 proposed that name. The founder and date of the genus can be- 

 ascertained by referring to any monograph." 



It is obvious on a little thought that this paragraph assumes a 

 good deal more than the facts warrant. In the first place there 

 ceitainly will be no monograph of the species named pubescens ^ 

 and it is very possible that a monograph of the generic name 

 chosen may not exist. 



But it is perhaps allowable to look at these two trinomials from; 

 a slightly different point of view. Which tells the most truth i 

 H. malachroides, H. and A., implies that H. and A. would nov^ 

 choose, as we have done, the group Hesperalcea for this plant. 

 This we have no right to imply ; as a matter of fact they did 

 choose Malva, and this is all we know or should state. 



Of course, in all the preceding I have assumed that the purpose' 

 of a name is to convey from one person to another the idea of a . 

 thing, and on this hypothesis it seems to me that the conclusions' 

 arrived at are sound ; but I would not wish to be understood as- 

 desiring that a name should do no more than this. If it can con- 

 vey the history of the thing, well and good, as long as by trying' 

 to do this it does not entirely defeat its own purpose, as I think E 

 have shown Hesperalcea malachroides, H. and A., would do, 



C. MiCHENEB.- 



San Francisco, Oct. 7. 



Notes on the Saturniidse, or Emperor and Atlas Moths. 



Although the family Satumiidm comprises the largest and: 

 some of the handsomest of all the Lepidoptera, it is still very im- 

 perfectly known. The larvae are mostly gregarious, and feed oie 

 trees. Many of them form cocoons, which are attached to the- 

 branches of the trees upon which they live, while others (at 

 least in South Africa) are said to pupate in the ground. I am^ 

 not certain whether it has yet been ascertained whether this lat- 

 ter habit has been proved to be peculiar to certain species or 

 genera, or whether the same species may form its pupa in dif- 

 ferent vvays, according to circumstances. 



There is doubtless a much greater variety of these insects in- 

 tropical countries than we are at present aware of. Many of the- 

 most remarkable species are only received singly, and often re- 

 main unique in our collections for years. Collectors rarely have 

 an opportunity of rearing them from the larvae, even if they 

 should meet with a brood, and many species probably feed on 

 lofty trees, quite out of reach, while the perfect insects are noc- 

 turnal in their habits. Many of the larger, and especially the 

 domesticated species of SatumiidcB from which silk is obtained ia- 

 India, China, and Japan, vary very much, and this is another 

 obstacle to their successful study. Many of these domesticated 

 breeds, and the various wild or semi-domesticated forms allied to» 

 them have been simply named, and not described; or perhaps 

 only the food-plants and localities have been indicated. These 

 useless names find their way into our collections and from thence- 

 into our lists and papers, and form a wholly unnecessary element 



