In this paper, we examined strength development at a polymer-polymer interface in terms of the dynamics and statics of random-coil chains. Interdiffusion of chain segments across the interface was considered to be the controlling factor for tack and green strength of uncured linear elastomers. This concept is similar to that proposed by Voyutskii and differs markedly from contact theories as proposed by Anand. In our approach, time dependent wetting first occurs followed by interdiffusion. Increasing contact pressure and temperature should promote the establishment of molecular contact (wetting) at the interface up to a saturation point of complete wetting. However, interdiffusion is retarded by increased hydrostatic pressure and enhanced by temperature in the usual thermally activated manner. The effect of pressure on diffusion is to reduce the volume available for the “hopping” process of segmental motion and subsequently decrease the self-diffusion coefficient. This effect is important in polymer processing where large hydrostatic pressures are encountered but is not very important in normal tack experiments where the contact pressures are much less than a kilobar. Thus, tack measurements should be dependent on pressure to some degree at short contact times but should be largely independent of pressure at long contact times. Increasing the test temperature increases the average interdiffusion chain segment length, l, but decreases the stress required to pull the chain out. Since l increases with temperature as l(T)∼exp −Qd/2kT and the stress decreases faster with temperature as σ(T)∼exp Qd/4kT, the tack evaluated at constant time will decrease with increasing temperature for the interdiffusion controlled process. The effect of molecular weight on tack at constant contact time, tc, is to increase the tack according to σ∼M1/2 for those molecular weights whose relaxation time t∞<tc and decrease the tack according to σ∼M1/4 for those molecular weights where t∞<tc. The tack should reach a maximum at a molecular weight corresponding to t∞=tc. The latter value could be used to determine the self-diffusion coefficient of the chains. At small contact times, the results might be complicated by wetting processes. The position of this maximum in σ versus M is relatively insensitive to the contact time, sincet∞∼M3 and M at the maximum will consequently increase as tc1/3. These predictions appear to be in agreement with much experimental data on tack and green strength reviewed and presented by Hamed and Rhee, but differ in many respects from their own interpretations of the same data. In conclusion, the most important results of our molecular dynamics approach to tack and green strength are as follows; (i) the tack, a, depends on t and M, as σ∼t1/4M−1/4, for t⩽t∞; (ii) the green strength, σ∞, depends on M, as σ∞∼M1/2; (iii) both σ and σ∞ depend on testing rate, ε˙, as σ∼ε˙1/2 for T≫Tg; (iv) the self-diffusion coefficient, D, depends on M, as D=A/M2 and can be measured mechanically using appropriate tack and green strength data. These results are not unique to elastomers, and most of them have been shown to apply to other polymer materials by our laboratory and some of them have also been investigated by Kausch's group in Lausanne using crack healing experiments on glassy PMMA.