Why lm doesn't work on a modified data.frame in R?

Question

When I change my data.frame I get an error and can't do the lm:

observation.not.i = area[-i, ]
fit.new.observation = lm(farm ~ land, data = observation.not.i)

Error is

Error in eval(expr, envir, enclos) : object 'land' not found 

I am using this in jackknife procedure as in the following:

r.jack = c(rep(0, 50))
y.hat = c(rep(0, 50))
for (i in 1:50) {
    observation.not.i = area[-i, ]
    fit.new.observation = lm(farm ~ land, data = observation.not.i)
    y.hat[i] = predict(fit.new.observation, data.frame(land.area=area[i, 3]))
    r.jack[i] = area[i, 2] - y.hat[i]
}

However when I just run fit=lm(farm~land,data=area) it works fine. Please let me know if you are aware of the cause of the problem.

Answer 1

y.hat[i] = predict(fit.new.observation, data.frame(land.area=area[i, 3]))

The fit.new.observation model is expecting a column called land in the newdata argument of predict, but the column is called land.area.