Page:Condor8(4).djvu/9

88 THE, CONDOR I VoL. VIII mother dressed in a night cap. Later on, when we saw them full grown, they got to be more owl like and dignified. An owl spreads terror among the small ground folk as a ghost among negroes. It is the owl's shadow-silent wings, his sharp, sound-catching ear and his night- piercing eyes that make him the superior of the mouse, the mole, the gopher and the rat. He fans over the field with an ominous screech that sets a mouse scam- pering to his hole, but his ear ha? caught the foot-steps; those wings are swift; those steel trap claws are always ready; his drop is sure, his grip is death. From an economic standpoint, it would be difficult to point out a more useful bird in any farming community. Like many other birds, the barn owl deserves the fullest protection, but man is often his worst enemy. Santa 3[onica, Cal. The Percentage of Error in Bird Migration Records BY WITMER STONE I N no branch of ornithology is it more difficult to obtain reliable data than in the study of bird migration. It is seldom that we see the actual migration in progress, and then it is but a small fraction of the movement that comes under our observation and that often under abnormal conditions. Consequently we are thrown back upon a comparison of the records of the oc- currence, or the dates of arrival and departure of birds at various points, in any deductions that we may nmke as to the direction and rapidity of their migratory flights. Without considering the possibility of error on the part of the observer there are many conditions which tend to impair the accuracy of such records, such as in- ability to be in the field every day (luring the migratory season, inability to cover the same amount of territory each day, and the recording by some observers of early stragglers which were not noted by others. To obviate the last, suggestions have been made to record the arrival of the bulk of the species; but this at once admits the personal equation into the problem, and I find that nearly all observers differ in their interpretatiou of the bulk arrival, especially in the case of species which are subject to a constant increase in num- bers from the first day that they are observed. The average date of arrival based on several years' observation is more accurate as a basis of comparison, but even then there is a large probability of error. Now most of the published tables of migration consist of the records of single observers at scattered points along the route of travel with generally large inter- vals between their stations. Scarcity of competent observers nmde it practically impossible to secure a large number of migration records from a limited area; but the wonderfnl increase in the pop.ular interest in bird study which we have recently witnessed has developed many able observers and renders the accumulation of this sort of data quite feasi- ble. It has been my privilege to study a series of local records of this sort kept at ' from 3o to4o stations each year, all within x5 or 2o miles of PhiladelPhia , by a corps of observers organized by the Delaware Valley Ornithological Club. These records are suggestive both in the apparent reduction of the percent- Read at the Twenty-third Congress of the A. O. U. in New York City, November, ?9o5.