One thing to note is that your A260/280 ratio is high for a DNA sample, suggesting that you have a contaminant.  The proper ratio of pure DNA is around 1.8.  RNA is higher, so you may have an RNA contaminant.  I assume you are using some sort of kit to purify the DNA and it's possibly been a while since the RNAse was added to one of the reagents.  RNAse is quite stable, but even it will degrade over time and result in RNA contamination in your minipreps.

I assume you are using the pedestal on the NanoDrop and that it's not the version with a cuvette.  As long as you are using the pedestal you are well within the linear range for the concentration of DNA.  It will be a good exercise to dilute the DNA and re-read, but I predict that your result will be similar.  (On the other hand, if you are using a NanoDrop 2000c and reading with the cuvette, I believe you are above the linear range).

You have concluded based on the agarose gel that your concentrations look similar, but that gel is pretty saturated.  I am guessing you loaded at least 1 or 2 ul of each sample.  When you make your dilutions to re-read on the NanoDrop, run a gel with some of those.  Aim for about the same amount of DNA you have in lanes 3 and 4 (which I am guessing is about 100-200ng).

electrophoresis and nanodrop provide quite different clues, I think. From my experience I'd rely on Nanodrop readings for concentration and purity and electrophoresis data for shearing. 

Hi! Nanodrop is not very good for measuring as consentrated DNA as you told. It is better for more dilute solutions. You could make dilutions and measure again. I do not think that you should trust the electrophoresis on this one, because it is more suitable for simply rough estimation of concentration. WBR Laura K.

I agree with Laura. My guess is your DNA is too concentrated that is why you are having different reading. Too much DNA in the nanodrop inhibit the machine to read the concentration. Try diluting your DNA 10x and then check again the concentration. With gel electrophoresis, some gel viewers have the ability to check the concentration of the DNA based on a control with known DNA concentration. YOu can check your gel viewer if it has this feature, you can try it if you want to! Goodluck!

I agree with the answers above that you're DNA concentration is too high for an accurate nanodrop reading. The other problem with the nanodrop is that if you have any RNA in your sample, even degraded RNA as nucleotides, those will also absorb at 260nm giving you an artificially higher reading. So it may be that one sample has more RNA than the other. I agree that by the gel they are nearly the same in concentration, and that is probably the correct answer. 

I agree with the previous answers. Dilute the sample and try reading it several times to see if the results are coherent. I sometimes find difficult to get reliable concentrations if I don't let nanodrop stabilize after turning it on. 

In general, nanodrop doesn't give always the most thustful result. If you want to be sure of your DNA concentration, it would be better to use a fluorometric method to measure the concentration. This will only measure dsDNA and in my opinion is more reliable. 

I agree with the general opinion. Maybe you should dilute the sample serveral times. In nanodrop the DNA concentrations are similar probably due to you arrived to a saturated value. 

One thing to note is that your A260/280 ratio is high for a DNA sample, suggesting that you have a contaminant.  The proper ratio of pure DNA is around 1.8.  RNA is higher, so you may have an RNA contaminant.  I assume you are using some sort of kit to purify the DNA and it's possibly been a while since the RNAse was added to one of the reagents.  RNAse is quite stable, but even it will degrade over time and result in RNA contamination in your minipreps.

I assume you are using the pedestal on the NanoDrop and that it's not the version with a cuvette.  As long as you are using the pedestal you are well within the linear range for the concentration of DNA.  It will be a good exercise to dilute the DNA and re-read, but I predict that your result will be similar.  (On the other hand, if you are using a NanoDrop 2000c and reading with the cuvette, I believe you are above the linear range).

You have concluded based on the agarose gel that your concentrations look similar, but that gel is pretty saturated.  I am guessing you loaded at least 1 or 2 ul of each sample.  When you make your dilutions to re-read on the NanoDrop, run a gel with some of those.  Aim for about the same amount of DNA you have in lanes 3 and 4 (which I am guessing is about 100-200ng).

Thanks for all the people.

What you are saying are quite resonable.

As I dilute sample 1 and sample 2 by 10x, 

their concentration become:sample1 80.5ng/ul, A260/280 2.091

sample2 140ng/ul, A260/280 2.121.

Just as pedicted by Dr.Whitacre.

And now I treat both sample with RNase, hopefully we will know the answer tomorrow.

But I still have a question.If it is contanminated by RNA, why we can not detect by gel electrophoresis?

I, too, was thinking about why we would not see the RNA on the gel.  There probably would not be much RNA and there is likely a fair amount of RNAse on you gel apparatus (from all the minipreps, which have RNase).  So, we would expect the RNA to be degraded during the run.  Also, treating with RNAse will de-polymerize the RNA, but the nucleotides will still be in there and absorbing UV light.

I would suggest taking your 1:10 dilutions and running 2ul of each of those on a gel.  I think you will then see the difference in concentration.

How precisely do you need to know the concentrations of these samples?  I think the concentrations you are getting from NanoDrop are reasonably close.  Maybe you can use them as-is?