cubation length because of "successive hatching." With sets 

 of eggs with which the successive hatching type prevails, 

 it is impossible to determine how long it takes to hatch each 

 egg in such a set unless the eggs be marked. Let us see 

 what can (and evidently does) hajipen in determining the 

 length of incubation with the robin, a species which lays 

 one egg each succeeding day until four or six are in the nest. 

 Occasionally the parents do not incubate steadily until all 

 their eggs are laid, in which case it is found that all the 

 eggs take almost exactly fourteen days to hatch, counting 

 from the laying of the last egg. If, however, the period 

 be dated as beginning Avith the laying of the first egg, it 

 would have to be estimated as being eighteen days, a pal- 

 pable ei'ror of four days. It is more common for this species 

 to have a set of eggs hatch irregularly ; it may be one on 

 the first day, two on the second day, and one on the third 

 day (of the hatching period), in which case no accurate 

 knowledge of the length of incubation could be gathered 

 without having marked the eggs for identification and in- 

 dividual study. Under these circumstances the first laid 

 eggs are partially incubated by the time the last ones are 

 deposited, causing the irregular hatching; and if the period 

 were counted as extending from the laying of the last egg 

 to the hour of the first hatching, the time elapsing would 

 probabl}^ be ten or eleven days, an estimate three or four 

 days too brief. I am convinced that many surprisingly 

 short incubation periods (as recorded in literature) are 

 much too brief, due to errors induced in the manner just 

 outlined. I am confident that the length of incubation of 

 the house finch is almost exactly fourteen days, but it could 

 be variously estimated as ten or eighteen days with different 

 sets of eggs if care were not taken to mark and carefully 

 identify the eggs as they are laid and hatched. Under such 

 conditions the larger the set of eggs, the longer or shorter 

 in days of error may be the estimate of the length of incu- 

 bation, errors (plus or minus) corresponding in days to the 

 number of eggs in the set, or to the number of days between 

 the laying of the eggs. The effect of partial incubation 

 when a set of eggs is being deposited results in mixed types 

 of hatching, an added source of conflict in the incubation 

 length data. 



On the other hand, it appears that some sets, embracing 

 several eggs, as w^ith the flicker (69-70), may have all the 

 eggs hatch at once (relatively), even though the early eggs 

 are apparently subjected to partial incubation. This may 

 be due to the possibility that the fresher eggs (later ones 

 laid) develop comparatively more swiftly (at the normal 

 rate) than do the older eggs (those laid first), resulting in 

 all breaking out at nearly the same time. The possibilities 



12 



