Because snags are important to forest wildlife as breeding, roosting a
nd foraging sites, resource managers who wish to maintain this compone
nt in forest stands need to be aware of snag fragmentation rates. Meas
urements were taken in uneven-aged pine-hardwood stands in southeaster
n Arkansas to determine fragmentation rates for hardwood snags 2 to 6
yr after stem injection with herbicides. Crown and hole condition of s
nags were also assessed. Pinus echinata Mill. and P. taeda L. were the
dominant overstory components and were undisturbed. Quercus spp, acco
unted for 91% of hardwoods >25 cm dbh. Since small diameter snags dete
riorated first, snag diameter distributions changed from uneven-sized
to even-sized structure as time since mortality increased. Within 3 yr
of injection, 57% of snag boles had broken below crown height. Number
of wildlife cavities per snag increased with time since mortality. At
6 yr after injection, 44% of residual snags had evidence of wildlife
cavities. Less than 50% of hardwoods <25 cm dbh were still standing 5
yr after herbicide injection.