184 



VARIATION IX ARUM MACULATUM 



counts he found 10% excess of the sinistral form, a result later con£rmed by Armitage 

 (1921). Further results are given in Table 3, and the help of obser\-ers is gratefully 

 acknowledged. 



The total figures show the differences to be highly significant. It may be noted 

 that there is no evidence of any clmal variation in this character. 



In some pohploids there is a relationship between pollen grain size and chromosome 

 number. Pollen from over a hundred samples of A. maculatum from over twents'-five 

 localities was measured but it was all of similar size. The pollen of plants from Denmark 

 (2n = 28) has the same diameter (41-5 zz -024; a = 1-7) as that of the British plants and 

 so also has the pollen of A. italicum Mill, and A. neglectum (Townsend) Ridl. 



3. Fruit varl\tiox 



This has been investigated by Colgan (1911) who compared spotted and unspotted 

 plants, and Salisbury (1942\ The following data are based on further counts of several 

 hundred fruits just before ripemng. The data may be presented and compared in various 

 ways. 



Table 4 



The aA'erage number of fruits in A. maculatum and var. immioculatum 



Total num.ber of Average num.ber Average number ripe 



fruits examined ovules per fruit seeds per fruit 



A. maculatum 1305 1-95 = 0-39 (a - 1-40) 1-36 = -036 (a = 1-17) 



var. immaculatum 2973 1-73 ± -024 (<j = 1-31) 1-lS = -021 (a = 1-09) 



The differences shown in Table 4 are statistically significant and further analysis 

 of the figures in Table 5 by the z- test shows that they are due to a tendency for A. macula- 

 tum to bear a larger num.ber of fruits with a high number of o\Tiles. 



Table 5 



The relation between the number of o\-ules per fruit and the total number of fruits 



Number of ovules per fruit 

 01234567S 

 A. maculatum 101 463 350 222 135 25 S 1 Data for '35, '40 and '41 



var. immaculatum 202 1215 926 473 120 33 3 1 



Xumbcr of ripe seeds per fruit 

 0 1 2 3 4 5 

 A. maculatum 163 549 161 14S 19 3 Data for '35, '36, *40 and '41 



var. immaculatum 396 1608 63S 119 11 — 



The data for Tables 4 and 5 are based on several years' figures, but if data for separate 

 years are analysed, they show that ni some years there is no difference between the per- 

 centages of ripe ovules per fruit m the two forms. Thus the difi"erences shown in Table 4 

 are really due to a large difference in one or two years. 



Table 6 shows no significant statistical differences in 1936, 1940 and 1941, but 



Table 6 



The number of fruits and o\-ules in A. maculatum and var. imm.aculatum for 1935, '36, '39, '40 and '41. 





Number of fruits 



Number of ripe seeds 



Year 



maculatum 



immaculatum 



maculatum 



immaculatum 



1935 



306 



600 



679 



930 



1936 



143 



824 



228 



1178 



1939 



376 



507 



368 



598 



1940 



208 



803 



212 



755 



1941 



341 



473 



435 



529 



