Two field cpcnmcnts were conducted at the Research and
ExpcrimcnLal Centcr of Moshtohor. Faculty of Agriculture in I991 and 1992
seasons. Three planting dates JcarI (May Is). intermediate (June t1) and laie
(July جf)J, two maize varieties (Gua 2 and T.WC 310) and five nitrogen
fcnhiicr levels (30. 60, 90. 120 and ILSO kg N/fed ) sserc csaluaied. Strip plot
design with three replications in first season and four in the second one for each
planting date was used Varieties were randomly assigned in the vertical stٌps
and nitrogen fertilizer Icscls in the hori,ontal strips (‘onibined anal»is over
both ears was made for the three planting dates of single eperimcnts for each
season scparatcl. The obtained results can be snmmareicd as follows
Rcsponsc curse anadsis:
In first season. data of M.is planting showed that the best nىodel fitted
to thc icld data of vartet Gua 2 was quadratic plus plateau model. whereas
linear plus plateau model fit well to the )icld ofT W C Ib aruc1% In June and
July planting. the best model was lincar plus plateau for the Iwo tested varieties
hi second season, the best model lined lo the yield data was Inicar plus
plateau model for the two sarictics en Ma planting and also for Giza 2 varict in
June planting The quadratic model fit wcll to the icld ot’ T W (‘ lIP sarict In
Jeme planting and also for both sariettes in Jul planting
Stcpwisc regression anal3 sis:
The results iiidicaled thai ؟K 7%’ el Ili total s;ereahuon in the sield
could be lincarl related to the studied sacuibles W ‘)9% of the total içld
s artatioui could be attributed to the s :iriahlcs uiieludcd iiito the model and I 7t,%
could be duc lo ihe cliinin:ited sanables
I’lie sanables included to the neodçl could be arr;i.igcd according io

their relatuse importance mc.esuicd b% partial R2. in a deseendiueg ordenas