For perspective, look at this fall's other daily counts:

Date	Number of Butterflies
Sep 26	0
Sep 29	NC
Oct 3	30
Oct 6	8
Oct 10	55
Oct 13	69
Oct 17	35
Oct 20	1,197

(Counts begin before dawn and last 1 to 2 hours. We count roosting monarchs predawn, over a one to one and a half hour period across two levees covering about a quarter of a mile each. our survey covers the two levees east and west of the St. Marks lighthouse every Wed and Sat through the migration period, which runs from early Oct to almost Thanksgiving. The count begins before dawn (e.g., 6:30) and generally runs a hour or more depending on the number of monarchs we find clustered in the trees and bushes. We count the monarchs in each tree or bush, then capture as many as we can, put the captured monarchs in mesh cages or in glassine envelopes to place in a cooler, and, after all else is done, tag and release them. If we have a large number of monarchs, we'll take 30 males and 30 females back to the office where we assess their condition, and record their wet weight and wing length. This same procedure has been in place since 1988, except for a five year hiatus in the mid-1990s.)

October 3rd
The first ripple of migrating monarchs reached the St. Marks National Wildlife Refuge. Wednesday we counted 30 monarchs in a one-hour period on the two levees leading out from the lighthouse. We caught and tagged all we could, and upon release some headed southwest out over the Gulf of Mexico. We began our fall surveying a week ago, but yesterday was the first day we saw monarchs roosting in the trees (at 6:00 a.m.). There have been monarchs in inland areas through most of the summer and early fall, but this is the first time we've had evidence that they've roosted overnight in the trees on these two levees. The two levees, along with coastal islands along the Gulf, generally serve as overnight staging (roosting) areas for migrating monarchs.