Physiological, biochemical and structural roles of nutrients in question determines their concentration in defined tissues.... N for instance has structural role in chlorophyll, is incorporated into amino acids hence its huge presence in leaves and photosynthesizing plant parts.... 

What about measure biomolecule instead of visual parameter such as leave or fruit. HPLC analysis or spectrophotometer will give you more robust and reliable results and you could be able to compare these results with the literature. Molecule such as chlorophyll, gluthation, MDA, are good indicator of stress. Dry and wet weight, stem length, number of leave are also good to have.

Fungai Mukange presented a set of functional factors to which I would add transport mechanisms via the phloem and xylem, and environmental factors such as availability  in the growth medium. Transpiration plays a role in driving the movement of mobile chemical species such as nitrate and chloride.  The number of possibilities is vast and the literature is equally vast.  Geoffrey Marchal is correct in stating that physiological parameters are good research tools.  However,  if you want a simple tool that can be widely applied to crops in the field you might do a lot worse than observing the plants and the soil in the field and testing both the soil and the plants in a laboratory.  These observations are complementary, i.e. the combination is a more powerful tool for diagnosing nutrient disorders and their likely causes than the separate components.   

Aabid H. Lone, depende de la especie con la que trabajes. Se debría hacer un análisis si son plantas anuales y una análisis totalmente diferente si son perennes. En plantas anuales y dependiendo del objetivo de las mismas el fruto podría ser un indicador de sensibilidad. No obstante, para que el fruto llegue a ciertos niveles de de un nutriente hay una correlación directa con lo que ocurre en otros órganos, por ejemplo las hojas. Hay varios trabajos en que la mayor concentración de N muestran mayor actividad fotosintética y, si tenes mayor actividad fotosintética seguramente tengas mayores contenidos de nutrientes en el fruto. También deberías tener en cuenta que mayor crecimiento, debido principalmente a mejores condiciones ambientales, podría producir efecto de dilución en la concentración de nutrientes. Esto te podría producir mayor demanda de nutrientes y a su vez, encontrar menor concentración en la planta y sus órganos (frutos, hojas, etc). Si las plantas son perennes, tambien son diferentes las opciones, pero si son especies de hojas no caducas (las hojas persisten en el arbol todo el año), el mejor indicador de sensibilidad son las hojas. Hay varios trabajos que muestran rangos mínimos y máximos de concentración de nutrientes foliares, dentro de los cuales la especie puede obrtener el mayor crecimiento.  

Aabid, diagnosis of nutrient constraints is , perhaps , as old as the history of plant nutrition .We should keep in mind while implementing the diagnostic method , you can not use  different types of analyses according to functionality of the nutrients ( Please see the classification criteria for essentiality of nutrients  ) , and neither according to functionality of different nutrient , you can use different  plant parts for analysis. This will make the practitioner into BEATING AROUND THE BUSH .In order to diagnose the nutrient deficiencies , a large number of diagnostic methods are used . they comprise of ; leaf/tissue  analysis  using index plant parts, soil analysis  ( often both are used in complementarity with each other), functionality of nutrients in terms of metalloenzymes ( For example ,  peroxidase , dismutases , carbonic anhydrase ,  dehydrogenase , nitrate reductase , go on adding ...) , flower/inflorescence analysis ( Gives earliest possible time for grower to address the deficiencies. Fully implementable results are not available , still in a developing stage as a diagnostic tool ) , juice/fruit analysis( Used for more stable results , but index values for all nutrients do not work , insensitive to many of the nutrients. But again sampling is not a problem , offers serious constraint towards early diagnosis),  and  deficiency symptoms( Often leads to erroneous results , because of overlapping symptoms , for example , N vs S, Fe vs Mn ). Therefore , the best way is to use leaf analysis well supported by soil analysis with a background history of deficiency symptomology. Probably in years to come , we need  a early earning system for diagnosing the nutrient deficiency in real time manner . The comments of Paul , Fungai  are very useful to put the interesting discussion into right perspective.

Very interesting , but different plant parts have different sensitivities with respect to different nutrients  and they have differential sink capacity for  accumulation of different nutrients across crop phonological stages. Our best premise would be to find out an index plant part which can act as a index for nutrient constraints appraisal. And this will be based on another premise it is a valid index plant part for most of the crunch nutrients . For example , index leaves ( May be 5-8 month-old from fruiting/non-fruiting terminals  covering atleast 8-10% trees in an orchard  with a minimum 50-80 leaves....like that ... it could vary as crop and index plant part type )  for citrus , mango , pomergranate , flag leaf for banana, third mature leaf for pineapple...like that  ...there could be so many .... But our intention has to address plant nutrition in totality rather than nutrient specific plant part . Such kind of attempt will complicate our intention of addressing nutrient constraints diagnosis and their remediation all the more complicated. Of late, floral analysis was suggested so that a grower has enough time to correct the results of diagnosis  within the same crop season . As Paul highlighted some physiological parameters can be very effective . In this context , metalloenzymes could be one area very sensitive and diagnosed earliest in crop phonological stage . But complications and ultra sensitivity of the analytical techniques  , make them least preferred by the researchers as well as practitioners . There could be leaf analysis complimented by soil analysis still remains unchallenged with regard to practical application in field.