FLOHAL VAUTATION J^ VERONICA PERSTCA 355 



bf the normal type, with the empirical formula K^C*A^Gr^. Three 

 further stages in the reduction of the flower have been observed by 

 me : IV, suppression of the anterior petal ; V, incomplete fusion of 

 the two anterior sepals ; VI, replacement of the two anterior sepals 

 by a single anterior one, the resultant flower having the formula 



Two additional stages may be conceived : VII, fusion of the two 

 "anterior" (originally lateral) petals; VIII, suppression of the anterior 

 sepal, the final result being a flower dimerous in all four whorls. 



A different type of reduction from the normal arrangement was 

 exhibited by tlie jiseudo-dimerous corolhis observed by me, in wliich 

 the empirical formula K'^C^A^G^ had apparently been attained by 

 the fusion of the anterior corolla-lobe with the lateral ones. 



Further investigations into the floral variation of V. pei^sica might 

 be made on the following lines : — 1, experimental cultivation in rich 

 leaf-mould, stiff clay, and sandy soil, respective!}'' ; 2, examination of 

 larger numbers of flowers (say, 10,000) in each of several different, 

 localities ; 3, separate examination v/eekly during the flowering- 

 season of 1000 flowers from a particular locality, in order to note any 

 differences in the ratio between reversionary and progressive abnor- 

 malities. It seems not unlikely that the percentages of reduced 

 flowers may be greater towards the beginning and the end of the 

 flowering-season than in the middle. 



^THE GENUS PTYSSIGLOTTIS. 

 By Spencer Le M. Moore, B.Sc, F.L.S. 



In the Flora of the Malay Peninsula, on p. 900 of the Gamo- 

 petalous portion (1907), C. B. Clarke published as a new genus under 

 the name of Leda certain Asiatic AcanthacecB assigned by Nees 

 (DC. Prodr. xi. 379) to Leptostacliya, and afterwards referred to 

 JDiantliera by Bentham and Hooker (Gen. Plant, ii. 1114). These 

 plants were very like Justicia in structure, the chief difference being 

 the absence in them of a spur from the lower or from both of the two 

 anther-cells. In his diagnosis of Leda Clarke describes the pollen as 

 ellipsoid, which is a Justicia character ; but my own observations, 

 backed by Clarke's own drawings in the Kew Herbarium, show plainly 

 that the pollen is globose with rarely a very slight difference only 

 in the two diameters. These pollen-grains are almost always quite 

 smooth (those of one species being minutely tubercled), and are 

 provided with two pores and very faint longitudinal banding near 

 the pores. In these characters all the species marshalled in this 

 paper agree. 



But by a curious oversight Clarke failed to perceive that in pro- 

 posing the new genus Lindau had anticipated him by more than ten 

 years. Describing his genus Strophacanthus (Engl. & Prantl. Nat. 

 Pflanzenfam. iv, 3 h, p. 344, 1895) the German botanist expressly 

 mentions as the members of it three Indian species, viz. Justicia 

 collina T. And. {Diantliera collina Clarke), Justicia dichotoma Bl. 



