SPAD readings are simple reflections of colour intensities and thus help to estimate the amount of chlorophyll content of a given plant only when a correlation and regression equations are available in that particular plant across the leaves of different colour intensities. This colour intensities of N deficient and sufficient plants are required to be carried out in firsthand and necessaqy correlation and regression equations are to be devised before recommending N requirement of crops based on SPAD readings.  

The SPAD measures the transmittance of the leaf, that is why I would perform measurements only with dry leaves. An other aspect is the soil water content. In dry soil the plant is not able to take up nitrogen, even if the soil is well supplied. Moreover, sulphur defiency leads to the same symptoms as N deficiency. So you will probably get wrong results for fertilising recommendations. I worked with a similar instrument, identical in construction. Here the measurements in cereals were correct because of a genotype-specific green colour of the leaves.

Concerning the specific effect of Mg on SPAD measure you can find usefull informations in:

- http://www.tandfonline.com/doi/abs/10.1080/01904167.2014.888742?journalCode=lpla20#.VAhubmO0OCA

- http://www.actahort.org/books/984/984_17.htm

It seem the effect is quite low and limited to young leaves.

SPAD readings are simple reflections of colour intensities and thus help to estimate the amount of chlorophyll content of a given plant only when a correlation and regression equations are available in that particular plant across the leaves of different colour intensities. This colour intensities of N deficient and sufficient plants are required to be carried out in firsthand and necessaqy correlation and regression equations are to be devised before recommending N requirement of crops based on SPAD readings.  

With a well develop calibration equation develop using regression of spad reading and different crop N status, the yes. The spad could be used for determining crop N status. But, time need to be invested in developing the calibration equation for the specific crop.

No, SPAD measurements should not be used to determine crop N requirements. SPAD measurements are great for quantifying chlorosis, but they do not indicate which nutrient may be causing the chlorosis. That can be due to N, S, Mg, and several micro-nutrients. In addition, pathogens and non-biotic factors (such as drought) can induce chlorosis.

SPAD  chlorophyll readers were originally developed by Minolta  to assist  farmers about  assessing nitrogen status of rice crops.  It is also quite useful when applied to other small grain cereals (wheat, barley etc.) and grasses. To my knowledge in maize it may be inadequate, particularly in discriminating among high N rates. 

Of course, in general you should develop a calibration with at least a few over-fertilized  and unfertilized plots in the field. Conversely, the assessment of crop N status  via SPAD readings in dicots may be troublesome, as the plants are less responsive to N fertilization. In the past (1996) we publish a paper on SPAD readings and total leaf chlorophyll of four crop species. It should be on my ResearchGate page.