766 



SCIENCE. 



[N. S. Vol. XX. No. 518. 



reviewer (pp. 503-5, October 14, 1904) with 

 the greater part of wliose remarks he is in 

 cordial agreement, and wliose work he has 

 ever found to be characterized by the greatest 

 care and the fullest consideration of every 

 point. 



My endeavor has been throughout to clearly 

 show the species and to some small degree 

 their variation, vegetative, continuous and dis- 

 continuous. My reason for not publishing in 

 detail my observations on the influence of 

 environment on mode of growth of corals was 

 simply lack of means to pay for the necessary 

 illustrations. I kept careful records of a 

 considerable number of species for this pur- 

 pose, but on reconsideration now I am of 

 opinion that any account, or at least the first 

 account, of the same should be written while 

 still on the reef and capable of verifying any 

 point that may arise. The best way would be 

 to take the necessary physical observations in 

 three or four different environments contain- 

 ing the same species of corals, to carefully 

 chart the character and organisms of the same, 

 and to show photographically their variations. 

 The specimens, from six or eight square yards 

 of each, being subsequently mounted in the 

 positions in which they grew in respect to one 

 another and other organisms, would form 

 most instructive and unique exhibits in any 

 museum which had the funds and enterprise 

 to undertake such a work. The best locality 

 would undoubtedly be the West Indies, since 

 its broader physical conditions are, perhaps, 

 better known than those of any other suitable 

 coral-reef region, and because it is the home 

 round which coral literature is entwined. 



I used the term continuous in respect to 

 variation in its ordinary sense, excluding, 

 however, vegetative variation, that due to the 

 action of the environment on the organism 

 during its growth. I do not restrict the term 

 to series from one locality grading into those 

 from others, as it may be equally well seen in 

 series from one locality, or even from one en- 

 vironment of the same. Where the modes of 

 series from diverse localities differ in certain 

 characters from one another, as seems to me 

 frequently to be the case, and yet the indi- 

 vidual specimens of the different species 



merge, we would seem to have before us true 

 cases of natural selection acting on continu- 

 ous variation. Yet — and this is the point — 

 neither new species nor new genera seem to 

 result from this action. 



In attempting to adopt a classification more 

 in accordance with our present ideas than that 

 of Duncan, I found that I had in the works 

 cited by Mr. Wayland Vaughan less than a 

 quarter of the information required to turn 

 out a ' good job,' and less than a sixth of the 

 necessary specimens in Cambridge required 

 to complete the other three quarters. Under 

 the circumstances I treated each genus by 

 itself, satisfying myself first as to its validity, 

 merely adopting a well-known order for con- 

 venience of reference. I tried to lay the 

 foundations in the species themselves, and I 

 left the gross classification in the hope that 

 some one, who has the specimens — I know no 

 one more capable and with wider knowledge 

 of recent and fossil corals than your reviewer 

 — might be induced to consider the whole. 

 Further, in the present state of our knowledge 

 I found it hopeless to work out a scientific 

 classification, based, as I consider such must 

 be, on the polyphyletic origins of the so- 

 called astrseid, fungid and perforate groups. 



I have no doubt Mr. Wayland Vaughan is 

 correct in respect to the generic names to 

 which he refers. I have, however, considered 

 the identity of Mussa with Symphyllia and of 

 Cceloria with Maecendrina, and in spite of 

 Professor Verrill's and Mr. Wayland Vaugh- 

 an's opinions I still separate them; the an- 

 atomy and development of their polyps and 

 coralla should be compared and would settle 

 the question. J. Stanley Gardiner. 



Cambeidge, England. 



some further comments on the guatemalan 

 boll weevil ant. 

 In his reply to my remarks on the feasi- 

 bility of establishing the Guatemalan boll 

 weevil ant in Texas, Dr. O. F. Cook shows 

 how dubious are the claims for the much- 

 advertised eificiency of this insect. Clearly 

 there are two distinct problems involved in 

 the discussion; first, the establishment of the 

 boll weevil ant in the southern states, and 



