Currently, only test winter data are used to evaluate the winter hardi
ness of alfalfa (Medicago sativa L.) cultivars. A test winter is defin
ed as field conditions severe enough to kill nonhardy cultivars and to
cause varying degrees of winter injury to cultivars of intermediate h
ardiness. Due to the unpredictability of test winter occurrence, alter
native methods are needed for predicting alfalfa winter survival. This
research tested the effects of temperature stress, storage time, and
plant size, measured as the width of the root at the crown-root junctu
re, on a laboratory cold tolerance evaluation of field grown and harde
ned alfalfa plants. Plants were dug from the field in mid-November and
stored at -2 degrees C until subjected to cold stresses from 0 to -24
degrees C. Differences among entries were greatest at the -12 and -16
degrees C cold treatments. Crown injury score was positively correlat
ed with held fall growth (r = 0.91) and winter injury scores (r = 0.95
; P less than or equal to 0.01). Root injury score was correlated posi
tively with field fall growth (r = 0.87) and winter injury scores (r =
0.95; P < 0.01). Large plants (> 10-mm root diameter) had lower cold
injury scores than small plants (1- to 5-mm root diameter). Correlatio
ns between plant size and winter injury were greatest (r > -0.55, P le
ss than or equal to 0.05) at low freezing temperatures. We propose a l
aboratory test in which field-hardened seeding-year alfalfa entries of
a similar size are packed in dry vermiculite, exposed to temperatures
of -12 to -16 degrees C for 1 to 2 h, thawed, and scored after 3 wk o
f regrowth.