242 Mr. G. A. Boulenger on the Races and 



most a very few, when large series can be examined, or 

 without reference to the data available through the labours 

 of other investigators, is unfair to those who make use of 

 works written on such lines. Systematics, if scientific, must 

 take into full consideration the exceptional, aberrant, or 

 annectant specimens, so often passed over without a word, 

 though of so great an importance from the taxonomic an 1 

 evolutionary points of view. It does not matter if thereby 

 our definitions are obscured, the object to be attained is to 

 depict the true state of things in Nature. 



To the four forms which I have previously * distinguished, 

 as torma typica, var. ridibunda, Pall., var. lessoinr, Gamer., 

 var. chinensis, Osb., I have recently added a fifth, var. 

 saliarica j", founded on specimens obtained by Dr. E. Hartert 

 in the far interior of the Algerian Sahara (El Golea, Tedikel 

 oases'), a small race nearly related to the var. ridibuvda of 

 the northern parts of Algeria but differing in the shorter 

 libipe, constantly less than half the length of head and body 

 and not overlapping when the limbs are folded at right 

 angles to the body ; the membrane between the toes is very 

 deeply notched, so much so that many specimens may be 

 described as having the foot oniy three-fourths webbed. 



The Vomerine Teeth. 



I have never seen these teeth in two series on the round 

 or elliptic bony bases that bear them, as described and 

 figured by FatioJ. They form a single series, composed of 

 3 to 8 ; in exceptional cases I find only 1 or 2 teeth (speci- 

 mens of the typical foim from St. Malo, Brussels, and Basle). 

 Leydig§ gives the number 3 as normal, but he cannot have 

 examined many specimens, those on which he drew up his 

 description being probably mostly of the var. lessona?, as the 

 figure of the foot given in his book indicates, and this number 

 is very frequent in the variety in question, although it may 

 rise to 5. In 8 frogs c 1 the typical form from Basle I find 

 only 2 to 4 teeth, whilst in 35 from other paits of Switzer- 

 land, from France, ar.d from Germany I count 3 to 7, 4 to 6 

 being the usual number ; 1 have also seen a toothless 

 specimen from Vienna. In about 30 specimens of the var. 

 ridibunda from Germany and Austria I count usually 4 to 6 

 teeth ; 3 specimens have only 3, one has 7 on one side and 



* Proc. Zool. Soc. 1891, p. 374. and Taill. Batr. Eur. p. 270 (1898). 



t Nov. Zool. xx. 1913, p. 84. 



t Vert. Suisse, iii. p. 313, pi. v. fig. 7 (1872). 



§ An. Batr. Deutschl. p. 112, pi. iii. tig. 20 (1877). 



