Null point accuracy measurement

Please refer the attachment, i had tried this question as in attachment "answer 1 and 2" getting the same answer as in the question.

However, now i realized in the answer given after differentiate
lnx=lnl1-lnl2 the differentiate result become plus. I have think for a long time couldnt get the reason behind.

Since the answer is the same, there must be some explanation behind it.

Attached Images

	null point accuracy.jpg (28.1 KB)
	answer1.jpg (40.7 KB)
	answer2.jpg (59.9 KB)