458 



NA TURE 



[September 8, 1892 



till we left at 3 p.m. We went direct from Tarona to Sandakan, 

 and as we steamed past the north end of the island I counted 

 18 jets of steam and smoke on the plain where in the morning 

 there had only been two. The volcano itself was wrapped in 

 smoke, and there were heavy clouds of smoke hanging over the 



plain The Dutch controleur told us that they had 



already recovered 300 bodies, but that it was impossible to esti- 

 mate the total loss. He said the other side of the island was 

 worse, lava as well as mud having overflowed there, and that 

 whole villages were buried. No lives were lost in Tarona itself, 

 but forty men from there went into the jungle just before the 



eruption, and only one got back alive We got to San- 



<lakan at midnight on the i6th, and stopped a day there. The 

 eruption was distinctly heard at Sandakan, though it is nearly 

 500 miles from Sangir." 



THE WEST INDIAN FAUNA IN SOUTH 

 FLORIDA. 



T)R. C. H. MERRIAM has lately published a paper on 

 ^-^ " The Geographical Distribution of Life in North 

 America" (Proc. Biol. Soc. Wash., April, 1892), which should 

 attract attention on account of the important problems discussed, 

 and the interesting and somewhat novel views advanced. On 

 pp. 49-55 there is a review of the faunal relations of Southern 

 Florida, in which Dr. A. R. Wallace is severely criticized for 

 having stated that Florida is, from a biological point of view, 

 essentially North American, and totally distinct in character 

 from Cuba and the Bahamas, from which it is separated by only 

 a narrow strait. The phrase specially attacked is out of " Island 

 Life," as follows: "Between frigid Canada and subtropical 

 Florida there are less marked differences in the animal produc- 

 tions than between Florida and Cuba."i 



I well remember that some time ago, when I knew next to 

 nothing of the West Indian fauna, this particular phrase seemed 

 to me very erroneous. An American zoologist cannot fail to be 

 struck with the presence of a colony of West Indian forms in 

 Southern Florida, so distinct from the species and genera of the 

 United States. Following Dr. Merriam's enumeration, we see 

 nine genera of tropical birds, hundreds of tropical insects, a 

 dozen or more land shells, many plants, and so forth. It would 

 seem impossible to doubt that Southern Florida should be 

 referred to ttie West Indian faunal division in the face of such 

 evidence. 



But if we examine the matter from the point of view of a 

 West Indian, who is searching for a fauna in Florida, which is 

 identical, or nearly so, with that of the islands, things look very 

 different indeed. Dr. Wallace's reference was to Florida as a 

 whole, the term " subtropical " being used as descriptive of the 

 State, not of the southern coast only, as used by Dr. Merriam. 

 In the map given by Dr. Merriam, about nine-tenths of Florida 

 are coloured orange, to indicate that they belong to the Lower 

 Sonoran Region of the author. Now this, with the Upper 

 Sonoran, which stretches into Canada, north of Lake Erie, 

 forms the Sonoran, one of Dr. Merriam's primary divisions, the 

 distinction of which from the Tropical region he has so well 

 demonstrated. Furthermore, a large part of Canada is coloured 

 blue on the map, to show that it belongs to the Transition 

 Region between the Sonoran and the Boreal. Hence it appears, 

 from Dr. Merriam's own map, and the statements throughout 

 his paper, that by far the greater part of Florida is more allied 

 faunally to portions of Canada than it is to the West Indies, so 

 far exactly confirming the truth of Dr. Wallace's statement. 



This will no doubt be readily admitted by Dr. Merriam, who 

 bases his criticisms on the ground that Dr. Wallace had over- 

 looked the existence of a West Indian fauna along the extreme 

 south coast of Florida. We may, therefore, consider the evi- 

 dence whereby this limited portion of the State is placed in the 

 Tropical division. For convenience, we may allude to this tract 

 as Tropical Florida, using the term tropical to indicate the 

 climate rather than the fauna. 



In order to get at the necessary facts, I have compared the 

 birds of the regions under consideration, using Cory's '* Birds of 

 the West Indies " as a guide to the ornithology of the several 

 islands. 



Dr. Merriam says that "no less than nine" genera of 

 Tropical American birds inhabit Tropical Florida, and cites 

 ' This statement is qualified by a footnote in the new edition of " Island 

 Life," where the existence of some West Indian forms is referred to. 



NO. II 93, VOL. 46] 



nineteen species or subspecies of Antillean birds living in the 

 same area, but not further north. 



I find on examining and comparing the West Indian statistics, ^ 

 that no less than 51 genera of West Indian land birds fail to 

 reach Florida or any other part of North America. These 

 genera are as follows : — 



Munocichla; Cichlherminia ; Margarops ; Ramphocinclus 

 Cinclocerthia ; Leucopera ; Catharopera ; Microligea; Teretistris : 

 Glossiptila ; Laletcs ; Du/us; Calliste ; Spindalis ; Nesospingus ; 

 Phanicophilus ; Calyptophilus ; Saltator ; Loxigilla ; Melopyr- 

 rha ; Loximitris ; Sicalis ; Nesopsar ; Elainia ; Lawrencia ; 

 Blacicus ; Nyctilius ; Siphonorhis ; Hemiprocne ; Glaucis ; 

 Lampornis ; Eulampis ; A it hum s ; Thalurania ; Mellisuga ; 

 Doricha ; Bellona ; Sporadinus ; Priotelus ; Temnotrogon ; 

 Saurothera ; Hyetornis; Todus ; Picumnus ; Nesoceleus ; Ara ; 

 Chrysoiis ; Gymnasia; Pupornis ; Pegerhinus ; Xiphidiopicus . 

 Those printed in ordinary type appear to occur in the West 

 Indies only in the Lesser Antilles, 



The number of West Indian species not reaching Florida is 

 of course overwhelmingly great, but here the comparison would 

 be unfair, owing to the large number of representative species, 

 on different islands. In order to obtain a just estimate I have 

 therefore made a list of the land birds inhabiting Cuba which 

 do not occur in Tropical Florida, and the result shows one 

 family ( Todidce), 18 genera, and 52 species. There are also a few 

 sub-species. 



The list is as follows : — 



Icterus hypomelas. Myiarchus sagrse. B. 



Agelaius humeralis. * Blacicus cariboeus. 



,, assimilis. Tyrannus magnirostris. 



Sturnella hippocrepis. Antrostomus cubanensis. 



Quiscalus gundlachii. Cypselus phoenicolius. 



,, atroviolaceus. * Hemiprocne zonaris. 



Corvus nasicus. Calypte helenae. 



,, minutus. * Sporadinus riccordi. B. 



Pitangus caudifasciatus. * Priotelus temnurus. 



* Saurothera merlini. Petrochelidon fulva. 



* Todus multicolor. Viveo gundlachi. 



* Xiphidiopicus percussus. * Spindalis pretrei. 

 Centrums superciliaris. * Melopyrrha nigra. 

 Colaptes chrysocaulosus. * Pyrrhomitris cucullata. 



* Nesoceleus fernandinas. Euetheia olivacea. 



* Mimocichla rubripes. * Ara tricolor. 



,, schistacea. Conurus euops. 



Myiadestes Elizabeth. * Chrysotis leucocephala. B. 



Mimus gundlachi. B. Asio stygius. 



Polioptila lembeyi. * Gymna'^io lawrencii. 



Dendroica petechia (race Glaucidium siju. 



gundlachi) B. Accipiter gundlachi. 



,, pityophila. ,, fringilloides. 



* Teretistris fernandinse. * Regerhinus wilsonii. 



,, fornsi. Columba corensis. 



* Coereba cyanea. ,, inornata. 



Geotrygon caniceps. 



The two species of Columba are not definitely given as 

 Cuban in Cory's work but I believe they occur there. Species 

 marked with an asterisk are of genera not reaching Florida ; 

 species marked " B " also occur in the Bahamas. 



The Bahama Islands also have many birds that are not in 

 Tropical Florida, including some genera, as Doricha (two 

 species). 



It is thus apparent that, so far as the birds are concerned, the 

 arm of sea between Cuba or the Bahamas and the mainland has 

 been very efficient in preventing the mingling of two faunas, 

 although a limited number of species have crossed it. 



To give many other instances would unduly prolong this 

 letter ; but one may cite the land shells as a much more striking 

 case. The land mollusca of Cuba and Florida are almost en- 

 tirely distinct, the small number (about a dozen^) of West 

 Indian forms which have reached Florida is really surprising, 

 considering the favourable currents and the proximity of the two 

 areas. Cuba contains numerous generic and subgeneric types, 

 and hundreds of species, which have never reached Florida.^ 



On the other hand, even on the Florida Keys we get such 



' These might be modified in slight details by searching the most recent 

 literature, but Cory's work. (i88g) is very complete up to the time it was 

 published. 



2 Dr. Merriam cites 20 on Dr. Dall's authority ; but several of these are 

 not land shells, but belong to brackish or fresh water. 



3 Thus, Cuba has considerably over 200 species of operculate land-shells 

 which have not reached Florida. 



