IN THE TROPICS: III. 1*1 



Among the experiments with sweet peas phenomena of 

 remarkable interest are recorded. The following case is of 

 particular importance. 



The white-flowered strain Emily Henderson includes plants 

 with long and with round pollen. When the two forms were 

 crossed together plants with purple flowers and long pollen 

 were obtained in Fi (reversion in colour). 



In Fjj the proportion of coloured flowers to white was 142 : 

 1, that of long pollen to round 336 : 1 (probably Mendelian). 

 The coloured flowers included several different types. As re- 

 gards colour the phenomena are clearly of greater complexity 

 than those described by Cuenot in the case of mice, but it is 

 probable that the explanation is to be looked for along similar 

 lines. 



An apparently new class of fact appears in the association 

 of long pollen with certain colours and of round pollen with 

 others. This coupling was not complete, but in some cases 

 long pollen was found on plants of the colour usually affected 

 by round pollen. The defmiteness of the characters concerned 

 makes the partial coupling in this case a more remarkable phe- 

 nomenon than, for instance, the coupling found in the case of 

 peas between floral colour and time of flowering, which was 

 described in Part I. of this series. 



Somewhat similar phenomena are described in the case of 

 the floral colours of stocks. In these plants a complication is 

 introduced by the fact that the characters hoariness and glab- 

 rousness are associated with the colouring in a manner not 

 completely understood. 



Recently published experiments by Biffen on the crossing of 

 wheats and barleys show that simple Mendelian phenomena 

 occur over a wide range of characters in these species. The 

 evidence of these experiments does not appear to lend any con- 

 firmation to the elaborate formulae enunciated by Tschermak 

 for the purpose of describing the supposed results of crossing in 

 cereals. 



