Here are two possibilities: 1) It probably easier to measure density than strength. Additional measurement noise could mean the strength difference is not stat

Here are two possibilities: 1) It probably easier to measure density than strength. Additional measurement noise could mean the strength difference is not statistically significant while the density difference is. 2) The relationship between density and strength is probably not a simple 1:1 linear function. For example a 10% bone mass reduction may only result in a 1% strength reduction. Therefore one many be statistically significant while the other is. Likely is is a combination of factors like this. Things are rarely simple in the lab.