We compute analytically the small-scale temperature fluctuations of the cosmic microwave background from cosmic (super-)strings and study the dependence on the string intercommuting probability $P$. We develop an analytical model which describes the
evolution of a string network and calculate the numbers of string segments and kinks in a horizon volume. Then we derive the probability distribution function (pdf) which takes account of finite angular resolution of observation. The resultant pdf consists of a Gaussian part due to frequent scatterings by long string segments and a non-Gaussian tail due to close encounters with kinks. The dispersion of the Gaussian part is reasonably consistent with that obtained by numerical simulations by Fraisse et al.. On the other hand, the non-Gaussian tail contains two phenomenological parameters which are determined by comparison with the numerical results for P=1. Extrapolating the pdf to the cases with $P<1$, we predict that the non-Gaussian feature is suppressed for small $P$.
High-energy emission from gamma-ray bursts (GRBs) can give rise to pair echos, i.e. delayed inverse Compton emission from secondary $e^{pm}$ pairs produced in $gamma-gamma$ interactions with intergalactic background radiation. We investigate the dete
ctability of such emission with modern-day gamma-ray telescopes. The spectra and light curves are calculated for a wide range of parameters, applying the formalism recently developed by Ichiki et al. The flux depends strongly on the unknown magnitude and coherence length of intergalactic magnetic fields, and we delineate the range of field strength and redshift that allow detectable echos. Relevant uncertainties such as the high-energy cutoff of the primary gamma-ray spectrum and the intensity of the cosmic infrared background are addressed. GLAST and MAGIC may be able to detect pair echo emission from GRBs with redshift $lesssim 1$ if the primary spectra extend to $sim 10 ~ {rm TeV}$.
