24 Buried Weed Seeds 



to conceive any way in which srreat quantities of Verovica Tonrnefortii, 

 V . hederaefolid and Briissica seeds could have been imported on to New 

 Zealand field during or after the ploughing. In fact, it would have been 

 utterly impossible in the case of Veronica hederaefolia, as this jilant is an 

 exceedingly earh' annual, and by the end of June or earlier it has shed 

 its seed m situ and the plants have died down. The seeds are produced in 

 abundance but have no provision for carriage by wind or animals, so 

 that after shedding they become mixed in with the soil where they fall 

 and lie dormant till the next season. As New Zealand was under grass 

 till the autumn of 1915, and seedlings of V. hederaefoUa were plentiful by 

 February, 1916, the proof is conclusive that the plants arose from seeds 

 which had lain dormant in the soil since the grassing down in 1906. 



Although six of the nine main species had appeared by February, 1916, 

 no trace was observed of the most abundant of all, Alriplex patula, nor 

 of Euphorbia exigun nor oi Poli/f/onum avicularc. This, however, accords 

 absolutely with expectations and with the glasshouse tests. In the latter 

 experiments two or three seeds of each of the three species germinated 

 soon after the pans were set up, in September and early October, after 

 which no more seedlings appeared till the following March or April, when 

 germination occurred freely. Evidently the winter period is inimical to 

 germination in these cases, and consequently the field was bare of the 

 seedlings in February, althoiigh by the following August they were all 

 present in abundance. The species that were present on the field in 

 February kept up a constant succession of germination in the pans 

 throughout the winter, so that in this respect also the field and glass- 

 house observations correspond closely. When the field was examined 

 later in 1916, all the arable species were found that had appeared in the 

 pans, with the single exception of Arenaria serpyllifolia. This was appar- 

 ently very scarce, and if plants existed on the field they may have been 

 overlooked or confused with small plants of Stellnria media. The domin- 

 ance of charlock was very marked, but it may be suggested that this was 

 not necessarily due to the very large number of plants occurring per unit 

 area, but to the large size that individuals attain and to the showiness of 

 the flowers. This explanation may help to reconcile the disparity be- 

 tween the dominance in the field and the comparative fewness of the 

 seedlings raised in the greenhouse. i 



The number of viable arable seeds obtained from the New Zealand soil 

 samples is in reality enormous and represents about 17 milHon per acre, 

 calculated on the average of the samples taken. Such calculations are 

 not really worth much, but they serve to show that even after land has 



