136 



ures ( PI. V, f. i) leaves of Trifoliitm repens, in one of which the 

 middle leaflet is small and dwarfed ; in a second this leaflet is modi- 

 fied as an awn ; while in a third it appears as a stalked ascidium, 

 very similar to the one here figured. Two other points are of 

 interest in connection with the instance here figured. First, the 

 character of the almost sessile pitcher in the leaf that has two, 

 where the margins adhere only at their distal portions. Second, 

 the fact that the ascidia are supernumerary leaflets. If they were 

 removed the leaf would still possess the number of parts normal 

 for the species (7! repejis ?). 



Ascidia in Licorice. — So far as I have been able to find, 

 pitcher-formation has not been previously recorded for the lico- 

 rice {Glycyrrhisa glabra). By inspection of figure g it is seen 

 that one leaflet of the fourth pair from the base is a shallow 

 ascidium. 



Fasciation in the Honey-locust. — In his "Elements de Tera- 

 tologic Vegetale" (Paris, 1841, p. 149) Moquin-Tandon, speaking 

 of fasciation in woody dicotyledons, includes the fevier, or Gled- 

 itschia in his list. According to Penzig (Pflanzen-Teratologie, p. 

 407), Camus recorded fasciation in the variety inerniis (Anomalie e 

 varieta nella Flora del Modenese. Terza contribuzione Rendi- 

 conti della Soc. dei Naturalisti dei Modena. Ser. 111,3. 1886). 

 I have never seen this fasciation figured before. The specimen 

 shown in figure h was brought to me by Miss Jean Broadhurst. 



Tricarpellate Eiiglish Walnuts. — Multiplication of the number 

 of parts in the flower is of very common occurrence. Moquin- 

 Tandon (/. c, p. 354), however, called attention to the fact that 

 polyphylly of the gynoecium is more rare than suppression of the 

 organs, on account of the pressure of adjacent parts. Masters 

 (/. c, p. 363) adds that the later development of the carpels is 

 also a factor here. He gives a list of eighty-nine genera in which 

 supernumerary carpels have been observed, and states that the 

 phenomenon is most common in Cruciferae, Umbelliferae, and 

 Liliaceae. Among some English walnuts {^Juglans regici) re- 

 cently purchased at a store, two or three nuts were found to have 

 three carpels (see figure /) instead of the usual two. While such 

 a thing is common in plants, Jnglans regia is not included in the 



